Stewart Guthrie

Animal Animism: Evolutionary Roots of Religious Cognition

There is no fundamental difference between man
and the higher mammals in their mental faculties .
. . the tendency in [humans] to imagine that
natural objects and agencies are animated by
spiritual or living essences, is perhaps illustrated
by . . . my dog [which] was lying on the lawn
during a hot and still day; but at a little distance a
slight breeze occasionally moved an open parasol
. . . every time that the parasol slightly moved, the
dog growled fiercely and barked. He must
[unconsciously have felt] that movement without
any apparent cause indicated the presence of
some strange living agent.
Charles Darwin, The Descent of Man

Permeated by [divine] power is everything that
moves in the universe.
First mantra, Isa Upanishad


Introduction

Despite Darwin, most scholars still see religion as marking a sharp divide between
humans and nonhuman animals. Religion, in their view, involves symbolism, wishful
thinking, explanation, or some combination of these activities, of which only humans
are capable. Thus any explanation of religion begins and ends with ourselves. A few
writers, however, root religion in a biological matrix shared by other animals. A key
part of this matrix is that we and other animals all inhabit ambiguous environments
that we must scan for hidden agents. In scanning for such agents, we encounter false
positives: we think we see agents where none exist.
I hold, then, that nonhuman animals display the common denominator of
religions: seeing more organisation in things and events than these things and events
really have. Like us, other animals appear to attribute characteristics of life and agency
to the inanimate world. In this sense, other animals are animists. This is because we all
respond to perceptual ambiguity in a strategic way, produced by natural selection:
when in doubt about whether something is animate or intentional, or is the result of
action by something animate or intentional, we assume that it is. Because all
perception is ambiguous and because natural and human deceptions increase this
ambiguity, both we and other animals always must assume that there is more to the
world than meets the eye.
Previously I have argued that religion is grounded in a general perceptual and
cognitive strategy. I now am extending this argument by suggesting continuities and
similarities between religious thought and action on the one hand and nonhuman
animal perception and behaviour on the other. Far from requiring unique
explanations, religious ideas are generated by cognitive mechanisms that are not only
ordinary (Hume 1957 [1757]; Darwin 1871; Tylor 1871; Horton 1960, 1993; Guthrie
1980, 1993, 1997a, 1997b, 2001; Lawson and McCauley 1990; Boyer 1994; Barrett
2000) but also broadly distributed in the animal world.
Religion doubtless has uses functions that are typically human
(McCutcheon 1997), but its existence cannot be explained primarily by such uses. It
can be explained, however, as a byproduct of something else: a perceptual and
cognitive strategy. As it arises, religion may well serve various purposes just as do
other accidental products of evolutionary process and serving such purposes may
encourage it.
The fundamental question is, however, why religion arises. (Jensen, this volume,
remarks that this always has been the grand question; and Anttonen, this volume,
writes that for scholars, there is hardly a more serious task.) My answer is avowedly
reductive. I find it in religions continuities with other thought and action, including
those in nonhuman animals, and in features of the natural environment in which such
thought and action evolved. This approach agrees with recent assumptions in
cognitive science about the fit between minds and the environments in which they
have evolved, as well as with older assumptions in biology (e.g., Uexkll 1992 [1934]).
Cosmides and Tooby (1994: 103), for instance, write that because ancestral
environments caused the design of psychological adaptations, investigating
environments should guide exploration of our cognitive mechanisms. That is,
understanding the natural environments of our ancestors will help clarify how our
own cognition works. Such clarification aims at what Sperber (1996: 4) calls the
prototypical natural-scientific goal, namely, to discover some natural mechanism
that explains a wide range of phenomena in a testable manner.
What is at stake in my argument is not so much the question of what analogues to
religion nonhuman animals may have, as what the answer can tell us about what
humans have and why they have it. I have argued earlier (1980, 1988, 1993, 1996b,
1996c, 1997a, 1997b, 2001) that religion may best be considered cognitively, as a form
of anthropomorphism (and of the related phenomenon of animism). Ive argued
moreover that the cognitive processes producing anthropomorphism and animism are
deeply intuitive as this term is used by Sperber (1996: 89), for example, who
describes intuitive beliefs as typically the product of spontaneous and unconscious
perceptual and inferential processes; in order to hold these intuitive beliefs, one need
not be aware of the fact that one holds them, and even less of reasons for holding
them. This describes most anthropomorphism and animism very well indeed and
hence, in my view, most religious thought and action.
In contrast, however, Sperber himself (1996) and Boyer (1994 and this volume),
Pyysiinen (this volume), and some others regard religion as in conflict with intuitive
thought and action. Indeed the latter two scholars find the explanation for the
existence of religion in this very conflict. Thus although we all approach religion
cognitively, we do so from quite different directions. Our diversity, however, is not
unusual among theorists of religion, who, as Anttonen (this volume) notes, still are
quite heterogeneous.


Is Religion Uniquely Human?

The prevailing answer to the question whether religion is unique to us is yes (e.g.,
Lessa and Vogt 1979; Carey 1995; Atran 1995; Mithen 1999; Rappaport 1999). In this
view, religion is different from anything in nonhuman animals. It is diagnostically and
decisively human, and depends on capacities and preoccupations that set us off
radically.
One familiar analysis of religion (Feuerbach 1972; Freud 1964; Malinowski 1948)
that limits it to humans is that it consists of wishful thinking. Caught in various
existential traps, not least our knowledge of our mortality, we fantasize ourselves
rescued by gods and enjoying some comfortable post-mortem existence. Since only
humans know that they will die, and only humans imagine something better, only
humans have religions
This wishful-thinking account (which appears as early as the Greeks and as recently
as Stark and Bainbridge 1987) doubtless has some truth, but it appears insufficient.
For one thing, many religious ideas picture a world we would not wish for, with
demons, angry gods, or terrifying afterlives. Indeed Burkert, noting that many
religions are frightening, writes, to transmit religion is to transmit fear (1996: 30-31).
For another, we do not believe just anything we want. As Pinker (1997: 555) puts it,
people freezing to death dont comfort themselves with a belief that they are warm.
That is, even if all religious beliefs were comforting, we still would need to know why
they are plausible. Thus the wishful-thinking theory is weak, and an absence of
wishful thinking in other animals would not necessarily mean the absence of the basis
for religion.
A second analysis of religion, often called intellectualist or, more generally,
cognitivist, is that it constitutes explanation (Hume 1757; Tylor 1871; Horton 1993)
and, of course, only humans explain. Again there is some truth here. Religions often
do offer explanations, and these, as far as explanation is propositional, require
language and symbols. But, as Lawson and McCauley (1990) argue, explanation is
closely interwoven with interpretation and is hard to distinguish from it; and certainly
other animals interpret the world, since all perception is interpretation, as
Wittgenstein pointed out.
A recent cognitive account of religion that also seems to make it unique to
humans is Boyers (1994, partly adopted by Pyysiinen and by Anttonen, both this
volume). This account explains not how religious ideas arise, but why they persist. As
in Darwinian selection, they persist because of a selective advantage: here, their
memorability. Memorability is conferred by a combination of intuitiveness and
counter-intuitiveness. Ideas with this combination are variously supernatural, non-
natural, and extra-natural. As this account relies heavily on the notion of the
counter-intuitive, and as animals give little or no instruction (much less in
metaphysics), we might conclude once more that animals cannot share the common
denominator of religion.
Several problems with this theory arise, however, that are pertinent to my own
argument. The most basic is that because the reproductive success of ideas is
independent of the success of their bearersunlike the relation of genes to their
bearersthe theory is inconsistent with the Darwinian account of selection. Only by
conceiving of ideas themselves as life forms (as does Sperber, below) can this
inconsistency be avoided. But such a conception, giving life to ideas, is both reifying
and animistic and hence best avoided.
Another problem arises, as Talmy (1995:647) and Guthrie (1996a) note, because
Boyer describes the counter-intuitive components of religious ideas as generated
randomly. Which ideas persist and which do not is determined solely by the need for
some balance of intuitivity and counter-intuitivity. Thus, memorable non-natural ideas
should encompass an endless range of combinations. But instead certain features,
such as the frequent invisibility and intangibility of agents, in fact commonly recur
while virtually limitless others, equally counter-intuitive, do not occur at all.
A third, related problem is whether invisibility and intangibility are indeed
counter-intuitive or merely, as Talmy suggests, consistent with experience in natural
environments. Ethology and human experience alike suggest that the latter is the case.
First, animals in the wild not infrequently behave as though in the presence of unseen
agents (for example, even with no predator in sight, they often act cautiously though
occasionally, as in the chimpanzee rain dance, with bravado). This behavior is not
surprising, since invisible agents are common in nature and in human experience, and
intangible ones are not unusual (Guthrie 1993 and below). And it is, of course, in a
natural context that human intuitions also have evolved.
Further, the ethologist Uexkll (1992: 377) recounts instances of animals from
various phyla acting in Umwelten (subjective worlds of perception and action) that
possess phenomena which, however, are visible to the subject alone and are bound
to no experiences. Such phenomena include search images triggered in the absence
of any objective stimulus. One might call such phenomena imaginary, but Uexkll
calls them, together with the consequent behaviour, magical. For example, a pet
starling, which never had caught or even seen a fly, one day was seen to suddenly
rush toward an invisible object, catch it in mid-air, return with the object to its perch,
peck away at it with its bill as any starling will do with a captured fly, and finally
swallow the invisible thing . . . There was no doubt that the starling had had the
apparition of an imaginary fly in its Umwelt. Evidently the starlings whole world had
been so charged with the feeding tone, that even without the appearance of a sensory
stimulus, the functional image of fly-catching, which was in readiness, forced the
perceptual image to appear, and this released the entire action chain. This experience
indicates that otherwise utterly puzzling actions by various animals should be
interpreted magically (Uexkll 1992:378).
Moreover, in human experience, people commonly act at a distance, as through
place markers, projectiles, land mines, and other artifacts. This produces what Gell
(1998) calls the distributed person: the fact that persons, through their distributed
artifacts, typically are agents in various places at the same time. They need be neither
visible nor tangible to have a presence.
On the other hand, the invisibility and intangibility of most gods and spirits, like
those of most animals, are neither complete nor intrinsic but only partial and
contingent (Guthrie 1993). They are achieved by the Greek gods, for example, by
producing a screen of fog or smoke. Burkert (1996, like Guthrie 1993) notes that
dealing with gods we never see is little different from dealing with monarchs or
distant trade partners we never see. In addition, many gods, such as those of the
Hawaiians and Aztecs, not only have been perfectly visible and tangible but also have
looked very much like humans thus enabling Captains Cook and Cortez to be
mistaken for returning deities.
Reports of agents who are invisible and intangible may demand our attention. As
Burkert suggests, however, this probably is simply because these features make such
agents more powerful and sometimes more sinister than agents without them.
The same attention would be demanded by reports of camouflaged men in the woods
or reports of invisible microbes on ones food. No supernatural qualities are
required.
This conclusion avoids a potential ethnological problem, moreover, since, as
Hallowell (1960) and Saler (1977) argue, the very notion of supernatural is Western
and culture-bound. As Horton (1993) observes, applying this notion to traditional
African religions is no more useful than applying it to nuclear physics on the grounds
that physics invokes invisible entities and powers. Rather, Horton continues, both
science and religion begin with familiar models those of common sense and
adapt them to account for a broader range of phenomena by changing or dropping
certain features. The planetary model of the atom, for example, began with the model
of our solar system, but changed the scale and dropped such features as color.
Similarly, such religious models as those of ancestors begin with persons and lineages,
and modify or drop features such as embodiment. Gell (1998) similarly argues that
magical thinkers using principles of contagion and sympathy do not so much
contradict modern physics as simply do without it; and Saariluoma, this volume, also
finds nothing unintuitive in inferring the properties God has in the human mind,
culture, and existing religious texts.
A corollary of Hortons comparison of religious and scientific theories is that the
more abstruse versions of both are the province of professional scientists and
theologians. Most of us stick closer to home, where matter remains solid and gods are
neither totally Other nor totally disembodied. (Barrett 1999 and 2000 also makes this
point, distinguishing ordinary religious ideas from theological ideas. In this volume he
writes similarly that regardless of theological tradition, in non-reflective contexts,
concepts of gods conform to intuitive expectations . . . about all intentional beings.
For most people, for example, God is not omnipresent but can be in only one place at
a time.)
The third and last human capacity held to make religion unique to our species is
that for language and associated symbolism. Scholars who assume that religion
essentially is symbolism are a majority; just a few of the best-known include
Durkheim (1915), Turner (1967), Geertz (1996), and Rappaport (1999). For most
scholars, religious symbolism is a means indirectly to express, promote, or control
certain social relationships and our feelings about them (Anttonen 1996, 2000 offers
persuasive versions of this view); and for most of these scholars, the uniqueness of
humans is hardly in question. Rappaport (1999: 16) writes in his last book, religion
emerged with language. As such, religion is as old as language, which is to say
precisely as old as humanity.
Against the view that symbols are uniquely human, one might counter that
chimpanzees and other apes have rudimentary symbolism (e.g., branch-dragging to
suggest a direction of travel and occasional false warnings, which Rappaport 1999: 55
calls proto-lies). Moreover Rappaport, who sees ritual as the source of religion,
makes plain that ritual is not only human but is shared also by the birds, the beasts
and even the insects (1999: 25). As an ethologist writes, attempts to place humans
apart from and above nonhumans, have, in a sense, backfired. Comparative research
in animal cognition has demonstrated evolutionary continuity in many cognitive
abilities. It has also shown how connected humans are to other animals (Bekoff
1998: 176). Nonetheless, religion as such like most human institutions typically
does involve well-developed symbolism including language, and other animals do not
have this. Language, of course, makes a big difference. Among other things (as
Jensen, this volume, puts it) it permits us to imagine collectively.
Animal animism, then, is not a system of symbols but rather a set of non-symbolic
features of perception, cognition, and action that appear basic to religion and that are
present also in nonhuman animals. These features include, as noted, perceptual
uncertainty (Kahneman and Tversky 1982), a need to discover hidden agents, and an
overestimation of agency. Although these features are unsystematized and non-
symbolic in most nonhuman animals (the chimpanzee rain dance, which appears
cultural and systematic, may be a borderline case), they are crucial both to humans and
to nonhumans.




Religion in a Biological Matrix

A few scholars of religion have tried to bridge the gap between humans and
nonhumans. One is A.F.C. Wallace (1966). For Wallace, religion is mainly ritual, and
to understand it we must set it in the wider context of ritual in other animals.
Wallaces evolutionary rationale is similar to mine: one can regard religion as a special
case . . . of a more widespread pattern (1966: 217). However, his (functionalist) view
of religion as essentially ritual that promotes cooperation among humans encounters
the problem found in all functionalist accounts: one cannot explain how something
arises by the uses to which it may be put.
More recently, Walter Burkert (1996) also tries to place religions within a
biological landscape (1996: 65). Burkert sets this biological context partly by finding
analogies for religion in nonhuman-animal behavior. Many of his analogies are
unpersuasive, such as that between religious sacrifice and the abandonment by zebras
of one of their number to lions, or the abandonment by a lizard of its tail to an
attacker. These analogies (as Saler 1999 notes) are distant or coincidental.
Nonetheless, Burkert produces a substantive, family-resemblance account of
religion that is somewhat plausible. In his account, religion grows directly from innate
dispositions that we share with other animals, especially with other primates. Most
important are dispositions to deal with the world in general as though it were social
and communicative. For all animals, Burkert (1996: 156) writes, the world is
composed of signs and signals. Among humans, who attribute language . . . to
nature (p. 163), the abundant signs in nature turn into [voices everywhere] . . . as if
every being, everywhere, were telling a message (p.160).
What is lacking both in Wallace and in Burkert is a unified account of what makes
these (usually) unseen agents credible. Burkert (1996) does note crucially, in my
view that we have inborn tendencies to think and act in social and linguistic ways
vis--vis the world at large, and that the ambiguity of the natural environments against
which these tendencies play out provides fertile ground for them (p. 118).
Nonetheless, he concludes that the problem of explaining religion serious
communication with powers that cannot be seen (p.176) has no single solution.
Instead, he simply proposes that there are biological patterns of actions, reactions,
and feelings that stem from our ancestral contexts of evolution. Well and good. But
what were those contexts, and how and why do the resulting patterns of perception
include humanlike, but non-human, others?


Animism Explained

The diverse biological patterns to which Burkert points may be unified under the
more general heading of animism. Animism, of course, has meant many different
things. Even within anthropology, the discipline that effectively created the term, it is
ambiguous, sometimes meaning the belief (Evans-Pritchard 1965: 24-25) that not
only creatures but also inanimate objects have life and personality and sometimes the
belief that in addition they have souls. In Tylors famous formulation, animism is
the belief in spirit beings though exactly what spirit beings are remains unclear,
as Saler points out (1993: 88-93). At present, owing largely to its widespread use by
post-Tylorian comparative religionists, animism most commonly means a belief in
living, personal powers behind all things (Pals 1996: 24).
The varied meanings fit, I think, within animism as defined by Piaget and other
developmental psychologists, namely as the attribution of characteristics of animacy
to nonliving things and events. For the kinds of life that most concern humans, these
characteristics may include form, such as eyes and bilateral symmetry. (As Dennett
1993 notes, through much of human evolution such symmetry may have represented
the face of a predator gazing at us.) More important, however, is behavior, including
spontaneous motion, responsiveness, and orientation to a goal. In cognitive ethology,
for instance, a philosopher and an ethologist remark that it may be a mistake to
assume that animals perceive predators mainly by morphology. Instead, predators
may be conceptualised according to what they typically do (Allen and Bekoff 1997:
121). Similarly, Uexkll, having described a tame jackdaw that chose a series of
companions including Konrad Lorenz, a younger jackdaw, and crows, notes that
there is no uniform perceptual image for the companion in the jackdaws world.
Nor could there be one, since the role of the companion changes all the time
(1992:371).
For such situations, Uexkll suggests, we should posit that animals employ not
specific search images but instead search tones, which are more general. Now we
do not always look for a definite object with a single receptor image, but far more
often for one that corresponds to a specific functional image. Instead of a specific
chair, we look around for something to sit on, that is, for a thing that may be
connected with a certain performance tone. In this case we cannot speak of a search
image, but only of a search tone (1992:375; emphasis added). This, together with the
case of the jackdaw and the imaginary fly, suggests once again that an emphasis on
visibility and tangibility in agents, with its corollary of an expectation of specific form,
may be misplaced.
Animism, in the sense of the attribution of agency to objects that do not have it,
appears widespread among both humans and animals. I shall first address the
question why this should be and then cite examples. Because Ive offered a general
theory both of animism and of the related phenomenon of anthropomorphism in
other places (e.g., Guthrie 1993), the explanation here will be condensed. However, a
preliminary note about what my explanation is not may be in order. This is the
common notion that animism and anthropomorphism are projections, a notion
popularised by Freud and adopted (and attributed to me) by Boyer in this volume.
Projection has no place in my explanation. Indeed, the whole of Guthrie 2000 is an
argument that, as a psychological concept, projection is without merit. (Its principal
basis apparently is a folk psychology, dating back to the Greeks, of vision as a
projection of beams from the eye.) Rather than projections, animism and
anthropomorphism describe certain perceptions that we have decided, after the
fact, were mistaken. The perceptual stance that produces these mistaken perceptions
as occasional byproducts, however, is no mistake but a vital and unavoidable strategy.
In brief, we and other animals live in a perceptual world that always is ambiguous
(as one writer on primate behaviour puts it, nature cloaks herself in many modes of
unpredictability [Miller 1997: 312]). In this world, we need to distinguish, among
other things, what is animate from what is not. The world is ambiguous (though
normally were not aware of this) because even the simplest perception is an
interpretation or, as Gombrich (1973) puts it, a bet. (Quiatt and Reynolds 1993: 5,
writing of primates, note similarly that perception, for all species, is the interpretation
of sensations.) Interpretation, in turn, aims at significance or meaning. The most
meaning, finally, is in things that are alive, not inanimate. Hence we spontaneously
interpret a tapping at our window as a visitor, not a branch, and a tickling on our neck
as a bug, not a loose thread.
Interpretation may be urgent, for example when what is alive is something we
may want to capture [figure 1], or something that may want to capture us [figure 2].
Insert figures 1 and 2 about here
---------------------------------------------------------------------------------------------------------
Because ambiguity is chronic and time is short, we generally must interpret without
enough evidence to be sure. In this situation, our strategy is, better safe than sorry
(Guthrie 1993, 1997a, 1997b, 2001, forthcoming; Boyer this volume adopts the idea):
when in doubt, we assume that it is alive. An S-shaped object on a woodland path
might be either a stick or a snake, for example, but we tend automatically to see it first
as a snake. As Ristau (1998: 139) puts it, a fail-safe mechanism for most species
would be to interact with an unknown object as though it were animate, and probably
predaceous.

When we look at the world of animals, the ambiguity of perception is exacerbated
by deception. This deception includes various means to invisibility and intangibility.
These means are widely diverse and almost universal among wild animals, since if an
animal is seen, heard or smelt it is potentially in danger [and hence] deception as a way
of life occurs throughout the animal kingdom (Owen 1980: 9,17).
Visual deceptions alone include at least five distinct sorts (Owen 1980). These are
camouflage (which makes the animal look like its background, often with
countershading that destroys the appearance of solidity), colors that are diverse or
changeable in individuals (this makes forming a search image difficult), structures and
colors that divert (this distracts predators from vital parts) [figure 3], colors and
patterns that startle, and imitation of noxious animals or material [figure 4]. Because
of these deceptions, we and other animals often encounter others without knowing it.
As a corollary, we often judge that others are there when they are not. The world is, in
fact, full of invisible others, and we must always assume more than we see.
Insert figs. 3 and 4 about here

In addition to animals that are invisible, many animals are virtually intangible, for
one of three reasons. One is a diffraction grating producing a shimmering opalescence
that makes size and location difficult to judge (Hinton 1973). A second is swarming,
flocking, or schooling. This, too, presents predators with a target of indefinite
location. A sea lion trying to feed on a school of sardines, for example, may be baffled
by the constantly shifting pattern confronting it. A third kind of intangibility
coupled with invisibility until microscopes appeared is that of being very small, as
in microorganisms. One result of the invisibility and intangibility of microorganisms is
that until recently, most people saw contagious diseases, such as tuberculosis,
smallpox, cholera, and plague, as the work of intentional but invisible agents. As I
have noted earlier (Guthrie 1993), people usually think they detect gods not by seeing
them--though this also happens--but by seeing natural phenomena as the results of
their actions. (Some see AIDS, for example, as divine punishment.) Moreover,
recognizing invisible agents does not require sophisticated (or counterintuitive)
thought, as a study of preschoolers understanding of germs suggests: three-year-olds
recognized that the causes of illness may be invisible . . . . children will focus on a
specific invisible entity when judging a potential causal relationship within the domain
of biology (Kalish 1996: 99, 100).
Another kind of deception often occurs within groups of animals: misleading
communication, such as warning calls when no danger is present. Such deceptions
typically distract competitors. Shrikes living in mixed-species flocks, for instance,
occasionally give false hawk warnings when a bird of another species is about to seize
an insect, causing the other to flee for cover and leave the bug for the shrike (Gould
and Gould 1994: 134).
Primates also deceive, often with apparent intent. Indeed, primate intelligence
may be largely machiavellian, or devoted to deceiving, and avoiding being deceived
by, their fellows (Byrne and Whiten 1988; Whiten and Byrne 1997). A vervet monkey
on the losing side of a fight between troops may give a false leopard alarm, frightening
off the combatants and resetting the contest at zero (Cheney and Seyfarth 1990). A
baboon chased by another may stop and stare as though at a predator, halting the
pursuer (Byrne 1995: 125).
Chimpanzees, bonobos, and orangutans appear even more deceitful. Both wild and
captive chimpanzees have been seen leading others away from hidden food that was
known to the leaders but not to the led (Goodall 1972:96, Byrne 1995:132). A captive
chimpanzee, subordinate to another that was afraid of the dark, went outside after
dark, made banging noises and other strange sounds, and came back inside looking
scared. His erstwhile oppressor, now frightened, approached him for reassurance
(Savage-Rumbaugh 1988: 228).
Capacities for deceiving others are related to capacities for self-deception and for
play (Mitchell and Thompson 1986). Two captive chimpanzees developed a game in
which they pretended to attack an imaginary opponent in a cage, and a captive
bonobo has produced imaginary playmates (Savage-Rumbaugh 1988). The sign-using
orangutan Chantek signed cat and dog animals which frightened and fascinated
him when he wished to prolong or start a walk, acting as though seeking out a
hidden cat or dog (Mitchell 1993: 81).
Burkert, invoking work on primate deception to help understand religion,
writes:

The suspicion has been voiced repeatedly that religion is mainly trickery
and make-believe produced by those who profit from it. Forms of deceit
abound already at prehuman stages. . . . The unseen in particular can be
the object of manipulation. [Monkeys, for example] may avoid
confrontation by staring into a corner and voicing sounds of alarm, as if
reporting there is a monster in the corner (1996: 24-25; emphasis
added).

Burkert also says religion assumes humanlike beings whom we normally do not see.
He notes, with Lawson and McCauley (1990), that ritual refers through formulaic
acts to nonpresent partners. However, deceptions in themselves would be a grossly
insufficient foundation for the origin of religion. Even among monkeys the trick
cannot be repeated very often without being recognized . . . The point is that the
common world of language characteristically produces contents beyond any
immediate evidence (Burkert 1996: 25).
In human communication, recent researchers have noted the ubiquity [and]
sheer ordinariness of deception (Lewis and Saarni 1993: v) and maintained that it is a
normal consequence of communication and information (just as illusion is a natural
outgrowth of perception). Communication, deception, and play all appear interlinked.
Creative abilities in this realm constitute abilities to envision a world and an
alternative, what is present and what is not. These abilities are graded among animals
and seem present among vervets and even among shrikes.
One consequence of ubiquitous deception, in addition to the inherent
ambiguity of perception, is that it is perpetually difficult to tell what is living from
what is not (Guthrie 1993, chapter 2). This prevents us from easily and reliably
applying any domain-specific expectations we may have to any particular part of our
environment. Carey (1995) levels this point against the notion, shared by Sperber
(1994) and Atran (1994), that such domain-specificity can characterise folk biology
and indeed make such biology an innate core module of the human mind. She
notes, for example (1995: 275), that in one of the cultures that Atran has studied, the
Itza Maya, fungi and lichens are not considered alive. Raising the problem of
perception, she asks (p. 278), On what basis does the [Sperber-Atran] input module
categorize entities as animals? The folk-biology module will be useless unless the
cognitive system can identify the animals in the world.



Animism Among Animals

The attribution of characteristics of animacy to the nonliving and to plants appears
widespread among complex organisms. Because illusion is an inevitable concomitant
of perception, such attribution may be universal. Among invertebrates, known
instances of susceptibility to such illusion range from echinoderms to insects.
Regarding sea urchins, for instance, Uexkll remarks that a receptor image held in . .
. general terms can always give rise to mistakes. This has already been shown in the
sea urchin, in whose world cloud and ship are constantly confused with the enemy
fish, because the sea urchin responds in the same way to any darkening of the
horizon (1992:370). Various insects are fooled by insect-eating plants, such as the
Venus fly-trap; and some flowers, by resembling bees or wasps, fool these insects into
pollinating them as they attempt to mate.
Among vertebrates, evidence of hard wiring to detect signs of life, and evidence
of corresponding illusions, are abundant both from experimental settings and natural
settings. The possible signs of life include both form and motion. Regarding form, for
example, we and many other creatures are built to be extremely capable of detecting
eyes (Ristau 1998: 141) and we respond readily to anything that resembles them.
Ethologists have found sensitivities to eye-like displays in fishes (even larval ones;
Miklosi et al. 1995), iguanas (Burger et al. 1991), garter snakes (Bern and Herzog
1994), wild birds (Scaife 1974), domestic chickens (Gallup 1971), and human infants
(Morton and Johnson 1991). The eye displays need not be convincing to a human
observer: even marbles on the ends of two sticks suffice to alarm chickens (Gould
and Gould 1994: 135). This sensitivity also is exploited by many species, especially
those of insects and fishes, that display false eye-spots as a means of defense.
Similarly, sensitivity to bilateral symmetry, a feature of most mobile animals, has been
shown in animals ranging from insects to birds to dolphins and other mammals.
Humans often exploit the animal tendency to animate. Fishermen can catch
various fishes with nothing more than a bit of white rag on a hook. Hunters can lure
waterfowl and other birds even with crude decoys. To frighten parrots from an
orchard, I have set out a simple, weather-beaten plastic owl and found it mobbed
within seconds by blue jays, blackbirds, and a cardinal. Human-like scarecrows the
world over are effective even with only a sketchy resemblance to people. Inuit, for
example, are able to funnel caribou into ambush using scarecrows consisting of tall
rockpiles.
In natural settings as well, birds and mammals may treat inert matter as though it
were alive. Bekoff (personal communication) reports young coyotes mistaking sticks
for grasshoppers, and a coyote stalking a blowing sage brush. Hinton (1973) reports
birds mistaking twigs for caterpillars, and Marshall Thomas thinks many dogs treat
cars as though they were animate (1993: 12).
Primates animate at least as broadly. Menzel (1997: 231) writes, So-called social
behaviour patterns can be directed toward inanimate objects. Rhesus macaque
(monkey) infants form attachments to cloth-covered surrogate mothers; tamarins (a
kind of monkey) will groom a fur-covered object; an infant chimpanzee will threaten
an unfamiliar piece of food. Vervet-monkey infants give the aerial-predator call on
seeing falling leaves (Cheney and Seyfarth 1990), and, as noted, a young adult vervets
false terrestrial-predator call caused opponents to flee from a phantom leopard
(Gould and Gould 1994: 135). Macaques on Gibraltar have threatened an electric
fence. When an observer there accidentally stepped on an electric cable in the grass,
an infant macaque screamed at its motion, evidently taking it for a snake (Anne Zeller,
personal communication). A baboon being chased may, as noted, stop and stare at
nothing, apparently producing a phantom predator that deters the pursuer.
Chimpanzees, bonobos, and orangutans show the most varied animism. In
captivity, as noted, they all may produce phantom playmates or monsters (sometimes
to fool a fellow ape or a caregiver). The orangutan Chantek engaged in chase games
in which he would look over his shoulder as he darted about, although no one was
chasing him. He also signed to his toys and offered them food and drink. Like
children, Chantek showed evidence of animism, a tendency to endow objects and
events with the attributes of living things (Miles 1993: 49). A captive chimpanzee
also has directed alarms calls toward its own shed teeth. Its caregiver (Sally Boysen,
personal commnication) thinks it saw the teeth as beings that had made its mouth
hurt and bleed.
Finally and most tellingly, wild chimpanzees (Goodall 1975, 1992, personal
communication; Whiten et al. 1999) often respond to thunderstorms, to rapid
streams, and to waterfalls with the kind of display (shaking and dragging branches and
rushing about vigorously) that they use as a threat against predators and other
chimpanzees. Observers have reported this behaviour in six communities of African
chimpanzees, out of nine communities that have been closely studied (Whiten et al.
1999). Goodall and many other chimp-watchers think this behaviour is indeed a
threat directed toward these inanimate targets as though they were alive. The response
is both widespread and indiscriminable from those toward actual, natural agents,
visible or not. Although no one knows whether chimpanzees think of storms and
streams as supernatural, any hypothesis that they do so seems unnecessary. More
likely, as Pyysinen (this volume) puts it, the thunderstorm triggers the same reaction
as does a predator, without a representation of a . . . counter-intuitive agent being
involved.


Animism and Anthropomorphism in Humans

Most scholars (e.g. Piaget 1929) concerned with animism attribute it to children and
to people in small-scale (tribal or primitive) societies, and most scholars see
anthropomorphism the attribution of human characteristics to nonhuman things
and events as unrelated to animism and as a minor though lamentable problem in
human cognition (Mitchell et al. 1997 present a recent set of exceptions). In contrast,
I see animism and anthropomorphism as pervasive in human thought and action, and
as closely related, spontaneous over-attributions of organisation to things and events
(Guthrie 1993; 1997a; 1997b). Just as animism may be seen as one result of a better-
safe-than-sorry strategy of perception in an ambiguous world, anthropomorphism
may be understood the same way. The two often overlap.
Anthropomorphism has been the object of criticism for hundreds of years, yet
it continues to flourish. It flourishes in the arts and even the sciences (Kennedy 1992;
Guthrie 1993, forthcoming; Mitchell et al. 1997). It also flourishes in the spontaneous
perceptions of daily life, as when we hear voices in the wind or the plumbing, or see
some mechanism as resisting us. Animism is similarly widespread. Evidence of the
pervasiveness of anthropomorphism and animism, even in complex, industrial
societies, comes from many sources (Guthrie 1993). Here Ill only draw briefly on
developmental, experimental, and clinical psychology, and on commercial art.
Systematic work in psychology on animism and anthropomorphism began with
Jean Piaget (1929), who called them both animism and found them universal among
young children. Piagets basic finding that child animism and anthropomorphism
are spontaneous and ubiquitous has been disputed in some details, such as its
relation to his scheme of cognitive stages. Generally, however, it has been broadly
supported (e.g., Inagaki and Hatano 1987; Inagaki 1989; Ochiai 1989; Cherry 1992;
Berry and Springer 1993; Harris 1994; Poulin-Dubois and Hroux 1994; Carey 1995).
Harris (1994: 308) writes, for example, A long tradition of work on animism shows
that children extend psychological explanations to . . . rivers, clouds, and so forth.
Among the features that produce animistic and anthropomorphic responses,
motion appears preeminent, as my epigraphs suggest. (Indeed Aristotle made a
capacity for self-movement the chief criterion of animacy.) Berry and Springer (1993:
275), for instance, write that very different methodologies have shown that certain
movements by inanimate objects elicit attributions of animacy from preschoolers,
and Poulin-Dubois and Hroux (1994: 329) similarly find that over-attribution of
mental states characterises preschoolers responses to moving objects. Such
responses, of course, reflect the importance of motion. As Barry (1997: 46) writes,
movement perception is so essential to our being . . . because the visual system has
evolved to alert us to danger or to the presence of potential food.
We not only animate the inanimate, but we also anthropomorphise the animate
or the apparently animate (Guthrie 1980, 1993; 1997a; 1997b, 2001, forthcoming),
whether moving or not. As Gigenrenzer (1997: 275) writes, human intelligence
cannot resist [attributing] human social categories, intentions and morals [to] non-
humans. Carey (1995: 279) notes that infants attempt to interact socially with a
mobile that moves in response to a leg kick. Richards and Siegler (1986) write that
children over a wide range of ages teleologically attribute anatomical and other
features of organisms (plants, for example) to the purposes of those organisms.
More generally, Dennett (1987) posits an innate intentional or design stance with
which we interpret the world, Keil (1994: 251) sees both teleological and intentional
modes of construal as attitudes with which people are natively endowed, and
Kelemen (1999: 280) proposes that an innate promiscuous teleology makes humans
prone to systematic biases in their reasoning about the natural world. Most directly
to my point, Kelemen also writes that one of the best indicators that people are
compelled to reason in teleological terms is provided by the ubiquitous phenomenon
of religion. Adults propensity to view objects and events as purposefully caused by
intentional agents or gods is . . . prevalent . . . . [F]or all the profound differences
between religions and their associated mythologies, a common moral tends to underlie
most: objects and events have an intended purpose. Everything has a function to
perform within a contrived natural order of which humanity is a significant part
(1999:280). This closely parallels Piagets classic description of teleology in young
children. The phenomenon also supplies (as Guthrie 1993 notes) an answer to
Humes (1932:I, 157) quandary about the nature of our sense that the world shows
designa sense so strong as to underlie the Argument from Design, for Gods
existence. As a Hume scholar writes, there is an apparently universal propensity of
the mind to see design in natural order and an insistent feeling in most of us that
natural order springs from a designer (Gaskin 1988:127, 6; emphasis his). Kelemen
dubs this phenomenon Promiscuous Teleology, noting that it is not inherently
restricted to any category of objects (1999:290). Bacon (1960), of course, centuries
ago began a scientific revolution by pointing out that humans, but not nature or even
living things in general, have goals, and that the widespread human failure to
understand this constitutes anthropomorphism.
Equally important for my purposes is one major correction of Piaget, made by
various developmental and cognitive psychologists. The correction is that, contrary to
his claim that animism disappears by age twelve, it appears to persist throughout life
(Dennis 1953; Sheehan, Papalia-Finlay, and Hooper 1980-81; Tamir and Zohar 1991;
Cherry 1992, Kelemen 1999). Hauser and Carey (1998: 84) write that people generally
are primed, perhaps innately so, to take an intentional stance . . . toward a wide array
of moving objects. Similarly, Barrett (this volume, pace Pyysiinen, this volume)
writes that given only enough evidence to believe an object can willfully initiate its
own action . . . children and adults automatically attribute a host of human-like
psychological properties. To illustrate, when trying (with little success) to place small
magnetized spheres in a particular configuration, college students described the
marbles as having beliefs, desires, and even personality traits, accusing some marbles
of attacking others and being deliberately mischievous.
The pervasive tendency to animate and anthropomorphise, moreover, is not
limited to spontaneous interpretations such as those of young children and the college
students cited but appears as well in more considered interpretations, including more-
or-less scientific work. Of numerous examples (Guthrie 1993, esp. chapter 6), a
particularly relevant one is Dawkinss (1978) genes, which notoriously are selfish, and
his memes, which similarly are self-interested and which replicate. Even more
strikingly, Sperber (1996: 1) describes ideas as born in and as invading brains, as
propagating, and as having descendants. Whereas the Greeks saw the anima or
life force as the central causal principle for the behaviour of human (and other)
bodies, Sperber gives this role to ideas. The central theme of this book is quite
simple, he begins (p.1). Our individual brains are each inhabited by a large number
of ideas that determine our behavior. These determinative ideas not only are born,
live and die but also constitute families (pp. 81 and 83).
One might consider Sperbers terminology here as mere metaphor (on p. 2 he
calls it barely metaphorical) except that his central notion, an epidemiology of
culture, depends on it. Equally important, if, as Lakoff and Johnson argue (e.g., 1999,
echoing Nietzsche 1966), most human thought is metaphoric, mere metaphor is an
oxymoron. What matters is not whether a given model is metaphoric or not, but
whether it is useful or not. In the standard view which I share animistic and
anthropomorphic models are not useful (though the strategy of perception that
produces them as byproducts, i.e., scanning first for what matters most, certainly is).
That animistic accounts such as Dawkinss and Sperbers nevertheless are appealing
illustrates the continued vitality of animism among us all.
Other evidence of pervasive animism and anthropomorphism comes from
clinical psychology, for example in Rorschach testing. Respondents see ink blots
mostly as humans or parts of humans, and as certain animals such as bats and
butterflies (Beck and Molish 1967). Other animals come next, followed distantly by
plants and inanimate objects. A cross-cultural study (De Vos and Boyer 1989)
suggests that this pattern is widespread. Still other sources of evidence include
folklore (Thompson 1955), literature, and graphic art, in which personification and
other forms of anthropomorphism, as well as animism, are common worldwide
(Guthrie 1993, figure 5). In sophisticated art, these phenomena are, to be sure, often
self-conscious and sometimes manipulative, not spontaneous; but their pervasiveness
and diversity there nonetheless reflect their power.

Insert fig. 5 about here


Neither anthropomorphism nor animism in itself, of course, constitutes religion.
Rather, religion is a form of them that is systematised, symbolically elaborated, and
taken seriously (Guthrie 1980; 1988; 1993; 1996; 1997; 2001). In this form, they are
adapted to varied political, economic, military, and other social purposes.


Conclusion: An Evolutionary Framework for Explaining Religion

Earlier writers on religion as anthropomorphism (most notably Spinoza 1955, Hume
1957; Feuerbach 1972; Freud 1964; and Horton 1993; and recently in a vein similar to
my own, Wenegrat 1990) or as animism have disagreed on the nature of these two
phenomena and have produced no broadly convincing explanation of them. As a
result they have produced no broadly convincing theory of religion. Part of what has
been lacking, in my view, is a cognitive account of anthropomorphism and of
animism that addresses them in general rather than only in religion. I have aimed
much of my work at just such an account.
A related part of what has been lacking is an evolutionary framework for the
issues in question. Such a framework, still very much in the making, potentially can
link us to our animal relatives by joining cognitive science to ethology. Such a
framework would encourage us to see that in chimpanzees, for example, both the
ability to create an imaginary playmate or monster, and the ability to track other
chimpanzees through the forest by visual signs such as litter and broken foliage, are
the ability to imagine what is not present. It is no great leap to the ability, famous in
hunter-gatherer peoples, to see game from tracks and other traces. This ability
means putting together a world from indirect evidence. This daily-life activity, of
course, is the same in science as well (for example, in positing subatomic particles
from their bubble tracks) and in religion (for example, in the Argument from Design).
In the search for an explanation of religion, I believe, we have been beguiled by
symbolism and misled by a false sense of human uniqueness. As a result, we have
forgotten a vital need that we share with other animals: to interpret an ambiguous
world and to discover real agents hiding in it. In the course of discovering those real
agents, all of us inevitably think we see agents where, in reality, none exist.


References

Allen, C. and M. Bekoff. 1997. Species of Mind: The Philosophy and Biology of
Cognitive Ethology. Cambridge, MA: The MIT Press.

Anttonen, V. 1996. Rethinking the Sacred: The Notions of human body and
territory in conceptualizing religion. In The Sacred and Its Scholars. Comparative
Methodologies for the Study of Primary Religious Data, 36-64, edited by T.A.
Idinopulos & E.A.. Yonan. Leiden: E.J.Brill.

Anttonen, V. 2000a. Sacred. In Guide to the Study of Religion, edited by W. Braun
and R.T. McCutcheon, 271- 282. London: Cassell.

Atran, S. 1994. Core domains versus scientific theories: Evidence from systematics
and Itzaj-Maya folk biology. In Mapping the Mind, edited by L.A. Hirschfeld and
S.A. Gelman, 316-340. New York: Cambridge University Press.

Atran, S. 1995. Causal constraints on categories and categorical constraints on
biological reasoning across cultures. In Causal Cognition: A Multidisciplinary
Debate, edited by D. Sperber, D. Premack, and A.J. Premack, 205-233. Oxford:
Clarendon Press.

Bacon, F. 1960 (1620). The New Organon and Related Writings. Edited by Fulton H.
Anderson. New York: Liberal Arts Press.

Barrett, J. and F.C. Keil. 1996. Conceptualizing a nonnatural entity:
Anthropomorphism in god concepts. Cognitive Psychology 31: 219-247.

Barrett, J. 1999. Theological correctness: Cognitive constraint and the study of
religion. Method and Theory in the Study of Religion 11: 325-339.

Barrett, J. and A. Hankes. Forthcoming. Losing our marbles? The attribution of
intentional agency to inanimates.

Beck, S. and H. Molish. 1967. Rorschachs Test. Vol. 2. A Variety of Personality
Pictures. 2nd 3d. New York: Grune and Stratton.

Bekoff, M. 1998. Playing with play: What can we learn about cognition, negotiation,
and evolution? In The Evolution of Mind, edited by D. Cummins and C. Allen, 162-
182. New York and Oxford: Oxford University Press.

Bern, C. and H.A. Herzog, Jr. 1994. Stimulus control of defensive behaviors of
garter snakes (Thamnophis sirtalis): Effects of eye spots and movement. Journal of
Comparative Psychology 108: 353-357.

Boyer, P. 1994. The Naturalness of Religious Ideas: A Cognitive Theory of Religion.
Berkeley: University of California Press.

Boyer, P. 1996. What makes anthropomorphism natural: Intuitive ontology and
cultural representations. Journal of the Royal Anthropological Institute 2: 83-87.

Burger, J., M. Gochfeld, and B.G. Murray, Jr. 1991. Role of predator eye size in risk
perception by basking black iguanas (Ctenosuara similis). Animal Behavior 42: 471-
476.

Burkert, W. 1996. Creation of the Sacred: Tracks of Biology in Early Religions.
Cambridge, MA: Harvard University Press.

Byrne, R. 1995. The Thinking Ape: Evolutionary Origins of Intelligence. Oxford:
Oxford University Press.

Byrne, R. and A. Whiten. 1988. Machiavellian Intelligence: Social Expertise and the
Evolution of Intellect in Monkeys, Apes, and Humans. Oxford: Clarendon Press.

Carey, S. 1995. On the origin of causal understanding. In Causal Cognition: A
Multidisciplinary Debate, edited by D. Sperber, D. Premack, and A.J. Premack, 268-
302. Oxford: Clarendon Press.

Cheney, D.L. and R.M. Seyfarth. 1990. How Monkeys See the World. Chicago:
University of Chicago Press.

Cosmides, L. and J. Tooby. 1994. Origins of domain specificity: The evolution of
functional organization. In Mapping the Mind, edited by L. Hirschfeld and S.
Gelman, 85-116. Cambridge: Cambridge University Press.

De Vos, G.A., and L.B. Boyer. 1989. Symbolic Analysis Cross-Culturally. Berkeley:
University of California Press.

Dennett, D. 1987. The Intentional Stance. Cambridge, MA: The MIT Press.

Dennett, D. 1993 (1991). Consciousness Explained. Harmondsworth: Penguin.

Dennis, W. 1953. Animistic thinking among college and university students.
Scientific Monthly 76: 247-249.

Durkheim, E. 1915 (1912). The Elementary Forms of the Religious Life. New York:
The Macmillan Company.

Feuerbach, L. 1972. The Fiery Brook: Selected Writings of Ludwig Feuerbach.
Translated with an introduction by Z. Hanfi. Garden City, NY: Anchor Books.

Freud, S. 1964 [1927]. The Future of an Illusion. Garden City: Anchor Books.

Gallup, G.G., Jr. 1971. Tonic immobility as a reaction to predation: Artificial eyes as
a fear stimulus for chickens. Bulletin of the Psychonomic Society 8: 614-616.

Gaskin, J.C.A. 1988. Humes Philosophy of Religion. 2
nd
edition. Atlantic Highlands,
NJ: Humanities Press International.

Gigenrenzer, Gerd. 1997. The modularity of social intelligence. In Machiavellian
Intelligence II: Extensions and Evaluations, edited by A. Whiten and R. Byrne, 264-
288. Cambridge: Cambridge University Press.

Gombrich, E.H. 1973. Illusion and art. In Illusion in Nature and Art, edited by
Richard Gregory and Ernst Gombrich. London: Gerald Duckworth.

Goodall, J. 1972. A preliminary report on expressive movements and
communication in the Gombe Stream chimpanzees. In Primate Patterns, edited by P.
Dolhinow, pp.25-84. New York: Holt, Rhinehart and Winston.

Goodall, J. 1992. In the Shadow of Man. Boston: Houghton Mifflin.

Guthrie, S.E. 1980 A cognitive theory of religion. Current Anthropology 21(2):
181-203, with CA treatment.

Guthrie, S.E. 1988. A Japanese New Religion: Rissh Ksei-kai in a Mountain
Hamlet. Ann Arbor: University of Michigan Center for Japanese Studies.

Guthrie, S.E. 1993. Faces in the Clouds: A New Theory of Religion. New York and
Oxford: Oxford University Press.

Guthrie, S.E. 1996a Theories of religion. Review essay on The Naturalness of
Religious Ideas: A Cognitive Theory of Religion (Pascal Boyer) and The Heathen in
His Blindness...: Asia, the West and the Dynamic of Religion (S.N. Balagangadhara).
American Anthropologist 98: 162-163.

Guthrie, S.E. 1996b. The sacred: A skeptical view. In The Sacred and Its Scholars,
edited by T. Idinopulos and E. Yonan, 124-138. Leiden: E. J. Brill.

Guthrie, S.E. 1996c. Religion: What is it? Journal for the Scientific Study of
Religion 35:412-419.

Guthrie, S.E. 1997a. Anthropomorphism: A definition and a theory. In
Anthropomorphism, Anecdotes, and Animals, edited by R.W. Mitchell, N.S.
Thompson, and H.L. Miles, 50-58. Albany: State University of New York Press.

Guthrie, S.E. 1997b. The origin of an illusion. In Anthropology of Religion: A
Handbook, edited by S.D. Glazier, 489-504. Westport, Connecticut: Greenwood
Press.

Guthrie, S.E. 2000. Projection. In Guide to the Study of Religion, edited by Willi
Braun and Russell T. McCutcheon, 225-238. London and New York: Cassell.

Guthrie, S.E. 2001. Why Gods? A Cognitive Theory. In Religion in Mind, edited
by Jensine Andresen, 94-111. Cambridge: Cambridge University Press.

Guthrie, S.E. Forthcoming. Anthropomorphism. In Encyclopaedia Britannica

Hallowell, A.I. 1960. Ojibwa ontology, behavior, and world view. In Culture in
History: Essays in Honor of Paul Radin, edited by S. Diamond, 20-52. New York:
Columbia University Press.

Harris, P.C. 1994. Thinking by children and scientists: False analogies and neglected
similarities. In Mapping the Mind, edited by L.A. Hirschfeld and S. Gelman, 294-315.
Cambridge and New York: Cambridge University Press.

Hinton, H.E. 1973. Natural deception. In Illusion in Art and Nature, edited by R.
Gregory and E. Gombrich, n.p. London: Gerald Duckworth.

Horton, R. 1993. Patterns of Thought in Africa and the West: Essays on Magic,
Science, and Religion. Cambridge: Cambridge University Press.

Hume, D. 1957 [1757]. The Natural History of Religion, edited by H.E. Root.
Stanford: Stanford University Press.

Hume, D. 1932. The Letters of David Hume, edited by J.Y.T. Greig. Two volumes.
Oxford: Clarendon Press.

Inagaki, K. 1989. Developmental shift in biological inference processes: From
similarity-based to category-based attribution. Human Development 32: 79-87.

Inagaki, K. and G. Hatano. 1987. Young childrens spontaneous personification as
analogy. Child Development 58: 1013-1020.

Kahneman, D. and A. Tversky. 1982. Variants of uncertainty. In Judgment Under
Uncertainty: Heuristics and Biases, edited by D. Kahneman, P. Slovic and A.
Tversky, n.p. Cambridge and New York: Cambridge University Press.

Kalish, C.W. 1996. Preschoolers understanding of germs as invisible mechanisms.
Cognitive Development 11: 83-106.

Keil, F.C. 1994. The birth and nurturance of concepts by domains: The origins of
concepts of living things. In Mapping the Mind, edited by L.A. Hirschfeld and S.A.
Gelman, 234-254. Cambridge and New York: Cambridge University Press.

Kelemen, D. 1999. Beliefs about purpose: On the origins of teleological thought. In
The Descent of Mind: Psychological Perspectives on Hominid Evolution, edited by
M.C. Corballis and S.E.G. Lea, 278-294. Oxford: Oxford University Press.

Kennedy, J. S. 1992. The New Anthropomorphism. Cambridge: Cambridge
University Press.

Lakoff, G. and M. Johnson. 1999. Philosophy in the Flesh: The Embodied Mind and
Its Challenge to Western Thought. New York: Basic Books.

Lawson, E.T. and R.N. McCauley. 1990. Rethinking Religion. Cambridge: Cambridge
University Press.

Lewis, M. and C. Saarni. 1993. Lying and Deception in Everyday Life. New York:
Guilford Press.

Malinowski, B. 1948. Magic, Science and Religion, and Other Essays. Boston: Beacon.

McCutcheon, R.T. 1997. Manufacturing Religion: The Discourse on Sui Generis
Religion and the Politics of Nostalgia. New York: Oxford University Press.

Menzel, C.R. 1997. Primates knowledge of their natural habitat: As indicated in
foraging. In Machiavellian Intelligence II: Extensions and Evaluations, edited by
A.Whiten and R.W. Byrne, 207-239. Cambridge: Cambridge University Press.

Miklosi, A., G. Berzsenyi, P. Pongracz, and V. Csanyi. 1995. The ontogeny of
antipredation behaviour in paradise fish larvae (Macro porlus opercularis): The
recognition of eyespots. Ethology 100: 284.

Miles, H.L.W. 1993. Language and the orangutan: The old person of the forest. In
The Great Ape Project: Equality Beyond Humanity, edited by P. Cavalieri and P.
Singer, 42-57. New York: St. Martins Press.

Miller, G.F. 1997. Protean primates: The evolution of adaptive unpredictability in
competition and courtship. In Machiavellian Intelligence II: Extensions and
Evaluations, edited by A.Whiten and R.W. Byrne, 312-340. Cambridge: Cambridge
University Press.

Mitchell, R.W. 1993. Animals as liars: The human face of nonhuman duplicity. In
Lying and Deception in Everyday Life, edited by M. Lewis and C. Saarni, 59-89. New
York: Guilford Press.

Mitchell, R.W. and N.S. Thompson. 1986. Deception: Perspectives on Human and
Nonhuman Deceit. New York: SUNY Press.

Mitchell, R.W., N.S. Thompson, and H.L. Miles (eds.). 1997. Anthropomorphism,
Anecdotes, and Animals. Albany: State University of New York Press.

Morton, J. and M. Johnson. 1991. The perception of facial structure in infancy. In
The Perception of Structure: Essays in Honor of Wendell R. Garner. G.R. Lockhead
and J.R. Pomerantz, 317-325. Washington, DC: American Psychological Association.

Nietzsche, F. 1966. Werke in Drei Bnden. Vol. 3. Hrsg. von K. Schlecher. Munich:
Carl Hanser.

Ochiai, M. 1989. The role of knowledge in the development of the life concept.
Human Development 32: 72-78.

Pals, D.L. 1996. Seven Theories of Religion. New York and Oxford: Oxford
University Press.

Piaget, J. 1929. The Childs Conception of the World. London: Routledge and Kegan
Paul.

Pinker, S. 1997. How the Mind Works. New York: W. W. Norton and Company.

Poulin-Dubois, D. and G. Hroux. 1994. Movement and childrens attributions of
life properties. International Journal of Behavioral Development 17: 329-347.

Quiatt, D. and V. Reynolds. 1993. Primate Behavior: Information, Social Knowledge,
and the Evolution of Culture. New York: Cambridge University Press.

Rappaport, R.1999. Ritual and Religion in the Making of Humanity. Cambridge:
Cambridge University Press.

Ristau, C.A. 1998. Cognitive ethology: The minds of children and animals. In The
Evolution of Mind, edited by D.D. Cummins and C. Allen, 127-161. New York:
Oxford University Press.

Saler, B. 1977. Supernatural as a Western category. Ethos 5: 31-53.

Saler, B. 1993. Conceptualizing Religion: Immanent Anthropologists, Transcendent
Natives, and Unbounded Categories. Leiden: E.J. Brill.

Saler, B. 1999. Biology and religion: On establishing a problematic. Method &
Theory in the Study of Religion 11: 386-394.

Savage-Rumbaugh, S. and K. McDonald. 1988. Deception and social manipulation in
symbol-using apes. In Machiavellian Intelligence: Social Expertise and the Evolution
of Intellect in Monkeys, Apes, and Humans, edited by R.W. Byrne and A.Whiten,
224-237. Oxford: Clarendon Press.

Scaife, M. 1976. The response to eye-like shapes by birds. Parts I and II. Animal
Behavior 24: 195-206.

Sheehan, N.W., D.E. Papalia-Finlay, and F.H. Hooper. 1980-81. The nature of the
life concept across the life-span. International Journal of Aging and Human
Development 12: 1-13.

Sperber, D. 1994. The modularity of thought and the epidemiology of
representations. In Mapping the Mind, edited by L.A. Hirschfeld and S.A. Gelman,
39-67. New York: Cambridge University Press.

Sperber, D. 1996. Explaining Culture. A Naturalistic Approach. Oxford: Blackwell.

Spinoza, B. de. 1955. The Chief Works of Benedict de Spinoza: On the
Improvement of the Understanding; The Ethics; Correspondence. Translated with an
introduction by R.H.M. Elwes. New York: Dover.

Stark, R. and W.S. Bainbridge. 1987. A Theory of Religion. Toronto Studies in
Religion, Vol. 2. New York: Peter Lang.

Thomas, E.M. 1993. The Hidden Life of Dogs. Boston: Houghton Mifflin.

Thompson, S. 1955. Motif-Index of Folk-Literature. Revised and enlarged edition.
Bloomington, IN: Indiana University Press.

Turner, V. 1967. The Forest of Symbols. Ithaca: Cornell University Press.

Uexkll, J. von. 1992 [1934]. A stroll through the worlds of animals and men: A
picture book of invisible worlds. Semiotica 89: 319-391.

Wallace, A.F.C. 1966. Religion: An Anthropological View. New York: Random
House.

Wenegrat, B. 1990. The Divine Archetype: The Sociobiology and Psychology of
Religion. Lexington, MA: Lexington Books.

Whiten, A., J. Goodall, W.C. McGrew, T. Nishida, V. Reynolds, Y. Sugiyama, C.E.G.
Tutin, R.W. Wrangham and C. Boesch. 1999. Cultures in Chimpanzees. Nature
399:682-685.

Whiten, A. and R. Byrne (eds). 1997. Machiavellian Intelligence II: Extensions and
Evaluations. Cambridge: Cambridge University Press.

Young, R.M. 1985. Darwins Metaphor: Natures Place in Victorian Culture.
Cambridge: Cambridge University Press.