Theory and the Evolutionary Psychology of Religion
The International Journal for the Psychology of Religion, 22:231–241, 2012
Copyright © Taylor & Francis Group, LLC
ISSN: 1050-8619 print/1532-7582 online
DOI: 10.1080/10508619.2012.679556
THEORY
Attachment Theory and the Evolutionary Psychology of Religion
Lee A. Kirkpatrick Department of Psychology
College of William & Mary
More than 40 years after its inception, Bowlby’s theory of infant-mother attachment remains widely
accepted and highly influential across many areas and applications of psychology, including the
psychology of religion. As compelling as the theory is for explaining phenomena within its natural
domain, however, its explanatory scope is inherently limited: There are many aspects of religion that
it cannot, and should not be expected to, explain. From the perspective of contemporary evolutionary
psychology—with which Bowlby’s original theory has much in common—the attachment system
is one among many functionally domain-specific cognitive adaptations that populate our species’
evolved psychological architecture. Evolutionary psychology offers a valuable perspective within
which the attachment system can be seen properly as just one (important) piece of a much larger
puzzle—of psychology in general and religion in particular—as well as a powerful and generative
paradigm for identifying and fitting together the many other pieces that will be required if we are
to progress toward a comprehensive psychology of religion.
It has been a little more than two decades since I first proposed in print that attachment
theory had the potential to offer a powerful theoretical perspective for the psychology of
religion (Kirkpatrick, 1992; Kirkpatrick & Shaver, 1990). The theory has since generated a
considerable body of empirical research which, as illustrated by this special issue of The
International Journal for the Psychology of Religion, continues to grow in new and creative
directions (for reviews, see Granqvist & Kirkpatrick, 2008, in press). I of course find this very
gratifying, but it is John Bowlby, not I, who deserves the lion’s share of the credit: My own
contribution has been mainly one of recognizing a good idea when I saw it. Attachment theory
Correspondence should be sent to Lee A. Kirkpatrick, Department of Psychology, College of William and Mary,
P. O. Box 8795, Williamsburg, VA 23187-8795, USA. E-mail: lakirk@wm.edu
231
232 KIRKPATRICK
itself was already more than two decades old at the time, and to this day it continues not
only to be widely accepted but also to generate new research across many subdisciplines of
psychology. The Handbook of Attachment (Cassidy & Shaver, 2008), now in its second edition,
contains 40 chapters and weighs in at more than 1,000 pages. There are few theories in the
history of psychology that can boast such staying power.
There are probably many reasons behind the success and longevity of attachment theory, but
I suggest that one very important one is that Bowlby basically got it right. I believe that there
“really is” an attachment system—in the same way that there “really is,” say, a visual system
or an eating-regulation system—that reliably develops in all humans (and many other species,
though with differences in detail). I believe that this system can be understood as a suite
of information-processing algorithms or psychological mechanisms that operate essentially as
Bowlby described: for example, by attending to environmental information about the proximity
of the primary caregiver (attachment figure [AF]) and cues of potential danger, combining
this information with stored knowledge of previous experience with the AF, and motivating
attachment behaviors when the AF’s proximity falls below a desired set point. I believe that
the recipe for building this cognitive/emotional system is coded in the human genome, because
it evolved via natural selection as a solution to the adaptive problem of protecting helpless
infants from predators and other dangers faced in ancestral environments by maintaining
proximity between them and their primary caregivers. I believe, in turn, that parents (especially
mothers) are motivated to respond in particular ways to attachment behaviors by an equally
real parental caregiving system—another evolved cognitive adaptation designed by natural
selection in humans and many other species. I believe that the attachment system reliably
produces certain patterns of individual differences, at least in part as a function of systematic
differences in environmental inputs that cause the system’s set point to be calibrated differently
in different individuals, depending on such factors as the chronic presence of danger cues and
the perceived reliability of the AF in responding to attachment behaviors. In the first part of
this article I explain why I believe these things to be true, in a way that I do not believe the
claims of most other psychological theories to be true—in other words, why I think attachment
theory was (and continues to be) such a “good idea.”
However, attachment theory is not, and cannot be, a comprehensive theory of the psychology
of religion: It offers an explanation for, and a theoretical basis for developing new hypotheses
about, only a limited range of religious phenomena. As powerful as it may be within this
domain, attachment theory is unlikely to provide many useful insights about such important
questions as the influence of religion on prejudice and warfare, the prevalence and nature of
polytheistic beliefs systems (which likely dominated human thought until relatively recently),
or the nature and origins of such widespread religious practices as sacrifices, elaborate rituals,
and shamanism (to name just a few). The attachment system, I argue, is just one among many
cognitive-emotional systems that populate our species’ evolved psychological architecture,
so a comprehensive theory of psychology—and by extension, a comprehensive psychology
of religion—must necessarily include many, many other cognitive adaptations beyond the
attachment system. The crucial question, then, is: Where should we look for equally “good
ideas” about these other systems and their role in shaping religious belief and behavior? In
the second part of this article, I suggest that the answer to this question is contemporary
evolutionary psychology, for the very same reasons that make attachment theory such a “good
idea.” I then conclude by briefly illustrating some ways in which an evolutionary-psychological
ATTACHMENT AND EVOLUTIONARY PSYCHOLOGY 233
approach provides a powerful and novel perspective on some of the most long-standing
problems and issues in the psychology of religion.
WHY ATTACHMENT THEORY IS A ‘‘GOOD IDEA’’
The primary reason that I believe Bowlby’s theory to be basically “true,” in a way that I do
not believe most other psychological theories to be “true,” is this: In the first volume of his
Attachment trilogy, Bowlby (1969) developed and explained in detail not only what he thought
the attachment system might be and how he thought it might work, but why it is this way and
not some other way. His justification for hypothesizing the existence of an attachment system
and the principles by which it operates were not based, like so many other theories, merely
on generalizations from extant findings or intuitions conjured in an armchair. Instead, he went
outside the psychology and psychoanalysis of his day to draw upon such disparate fields as
control systems theory, ethology, and evolutionary theory, the combination of which led him
to an entirely new way of thinking about the nature and functional organization of human (and
other species’) psychology. In this sense Bowlby’s genius was not so much in the product of
his theorizing but the process by which he went about it—a way of thinking that was otherwise
to remain largely dormant within psychology until it reemerged nearly two decades later in
the form of evolutionary psychology (e.g., Cosmides & Tooby, 1987; Symons, 1987; Tooby &
Cosmides, 1990, 1992).
The extent to which Bowlby’s (1969) approach to human behavior anticipated contemporary
evolutionary psychology (EP) is rather remarkable. Specifically, the manner in which he
constructed his theory, piece by piece, illustrates several of the central defining features of
contemporary EP, including the following:
1. Bowlby began by acknowledging the utility of thinking about human behavior in the
context of animal behavior in general. He drew heavily upon ethology and comparative
psychology to place human behavior in this larger context. For example, he saw imprint-
ing behavior in precocial birds (Lorenz, 1957) as functionally analogous to human attach-
ment, and saw the importance of Harlow’s (1958) classic research on infant monkeys’
behavior toward artificial “mothers” in the laboratory. He recognized that explanations
of human behavior should be consistent with that of other animals’ behavior while
recognizing important differences across species.
2. Bowlby appreciated that evolved behavior patterns in all species, as a product of natural
selection, are organized around adaptive problems faced by that species throughout its
evolutionary history. In the case of attachment, the adaptive problem to be solved is
that of increasing the likelihood of survival of helpless infants from predators and other
dangers, by maintaining proximity between the attached infant and its AF. He noted that
the evolved imprinting system in goslings, per Lorenz, was designed by natural selection
to solve an adaptive problem similar to that solved by the attachment system in primates
but that the details of the evolved solution necessarily differ across species in light of
other ecological and biological constraints (e.g., that goslings, unlike human infants, are
locomotive from birth).
234 KIRKPATRICK
3. Following from the previous point, Bowlby recognized that…
