Content uploaded by William george Irons
Author content
All content in this area was uploaded by William george Irons on Apr 30, 2020
Content may be subject to copyright.
Content uploaded by William george Irons
Author content
All content in this area was uploaded by William george Irons on Apr 30, 2020
Content may be subject to copyright.
ARTICLES
Adaptively Relevant Environments Versus
the Environment of Evolutionary Adaptedness
WILLIAM IRONS
Manyresearchersbelievethatthecon-
cept of adaptation is useful for under-
standing the human mind and human
behavior.1–7 These researchers agree
that adaptations are design features of
organisms that evolved because they
enhanced fitness in ancestral environ-
ments. They see the psychological
mechanisms that make up the human
mind as evolved adaptations. Further,
they are convinced that these adapta-
tions are more likely to produce adap-
tive effects in environments similar to
ancestral ones. In other words, the
more similar the present environment
is to the ancestral one, the more likely
the adaptation is to confer the repro-
ductive advantage that led to its evolu-
tion. On the other hand, adaptations
are less likely to confer an adaptive
advantage in novel environments.
Despitethesesharedviews,theques-
tion of exactly how to characterize
these expectations has led to major
disagreement among researchers who
study human behavior and psychol-
ogy from an evolutionary perspective.
One group, whose members label
themselvesevolutionarypsychologists,
has dealt with this problem by elabo-
rating the concept of the environment
ofevolutionary adaptedness, (EEA).8–9
Other researchers, who are variously
labeled behavioral ecologists, evolu-
tionary ecologists, sociobiologists, or
human paleontologists, have tended
to question the value of this con-
cept.10–14
In this paper, I review and critique
the concept of the EEA and the associ-
ated evolutionary psychological view
that the human mind consists of many
specific-purpose decision-making
mechanisms rather than just a few
general-purpose ones. I then suggest
an alternative to the EEA concept that
I believe will better serve the purpose
of modeling the relationship between
adaptations and environments. I see
this concept as a more logical comple-
mentthan the EEA to the view that the
human mind consists of many specific
mechanisms. I refer to this new con-
cept as the adaptively relevant environ-
ment (ARE). The expression ‘‘relevant
environment’’ may also serve as a
shorter label. The key idea motivating
the ARE concept is that an organism
consist of a large number of special-
purpose adaptations, each interacting
with only a part of the organism’s
environment. Thus, when a particular
element of an environment changes, it
is likely to affect some adaptations but
notothers.Logically, this idea is closely
related to the idea that evolutionary
change is mosaic: In the course of
evolutionary change, some aspects of
organisms change while others re-
main the same. In order to understand
an adaptation fully at the proximate
level, we need to study its design, the
structure of its relevant environment,
and the interaction of the two.
Before proceeding, a word of cau-
tion is necessary regarding the label
evolutionarypsychology. The label has
both a broad and a narrow meaning.
In its narrow meaning, it refers to the
research program of scholars such as
Barkow,Cosmides,Symons,andTooby
who rely heavily on the EEA and asso-
ciated concepts and who insist that
others who do not share this emphasis
are not strict Darwinians or true adap-
tationists.2However, many writers use
the terms in a broader sense that
includes all recent attempts to study
human behavior and psychology in
evolutionary terms. Robert Wright’s
recent book, The Moral Animal,15 uses
the word in this broader sense.
THE ENVIRONMENT OF
EVOLUTIONARY ADAPTEDNESS
The EEA concept was introduced by
John Bowlby in his classic 1969 and
1973 studies of attachment and loss in
human infants and children.16–17
Bowlby defined a species EEA as ‘‘. ..
the environment in which a species
lived while its existing characteristics,
includingbehavioralsystems,werebe-
ing evolved, and . . . the only environ-
ment in which there can be any assur-
ance that activation of a system will be
likely to result in the achievement of
its biological function . . .’’ (p. 82).17
He identified the human EEA as ‘‘. . .
the one that man inhabited for two
million years until changes of the past
few thousand years led to the extraor-
dinary variety of habitats he occupies
today.’’(p.59).16 Bowlbydefinedadapt-
edness as contributing to the survival
of the species or populations. How-
ever, current evolutionary psycholo-
gistsnowdefineit,inlinewithcontem-
porary theory, as serving the inclusive
fitness of individuals. Aside from this,
they have not modified the concept
William Irons is Professor of Anthropology
at Northwestern University. He is currently
doing a meta-analysis of empirical studies
of the effect of status on reproductive suc-
cess in traditional societies. This project is
a continuation of his earlier work in which
hewasthefirstto present quantitativedata
onthis subject. Thisdata and hisinterest in
the subject of status and reproductive suc-
cess grew out of an ethnographic and de-
mographic study of the Yomut Turkmen of
Northern Iran. Irons has also published
theoreticalpapers on theevolutionary foun-
dations of morality and religion. E-mail:
w-irons@nwu.edu
Keywords: environment ofevolutionary adapted-
ness; adaptively relevant environments; evolu-
tionary psychology; human evolutionary ecol-
ogy; human behavioral ecology
194 Evolutionary Anthropology
greatly. Barkow, Cosmides, and Tooby
review the basics of this concept in the
introduction to The Adapted Mind,2an
edited volume designed to present and
explain the newly developing field of
evolutionary psychology to a wider
audience. They summarize the EEA
assumptions as follows: ‘‘. . . the
evolved structure of the human mind
is adapted to the way of life of Pleis-
tocene hunter-gatherers, and not nec-
essarily to our modern circum-
stances.... The few thousand years
since the scattered appearance of agri-
culture is only a small stretch in evolu-
tionary terms, less than 1% of the two
million years our ancestors spent as
Pleistocene hunter-gatherers’’ (p. 5).
One of the most vigorous advocates
of the EEA concept has been Donald
Symons. He first emphasized the con-
cept in his path-breaking 1979 trea-
tise, The Evolution of Human Sexual-
ity.18 In this book he used the label
‘‘naturalenvironment’’for what he and
other evolutionary psychologists later
came to refer to as the EEA. In charac-
terizing the ‘‘natural environment’’ of
humankind he says the following:
Large-brained hominids with ad-
vanced tool technologies have
existed for more than one million
years....Forover 99 percent of
this period humans lived in small
nomadic groups without domes-
ticated plants or animals. This
hunting and gathering way of life
is the only stable, persistent adap-
tation humans have ever achieved
. . . it is generally agreed that in-
sufficient time has elapsed since
the invention of agriculture
10,000 years ago for significant
change to have occurred in hu-
man gene pools....Humans can
thus be said to be genetically
adapted to a hunting and gather-
ing way of life . . .’’ (p. 35).
As a consequence, Symons believes it
is irrelevant to measure the reproduc-
tiveconsequenceofanyhumanbehav-
ior outside of a hunting and gathering
society. Symons has been especially
sharp in criticizing what he has la-
beled Darwinian anthropologists or
Darwiniansocialscientistsfor measur-
ing reproductive consequence and for
arguingthatbehaviorinsome contem-
porary nonforaging societies is adap-
tive.8Tooby and Cosmides share the
view that measures of current fitness
are irrelevant in both human and ani-
mal studies. They recently wrote, ‘‘It is
a surprising lapse in many excellent
evolutionary researchers’ thought (see
e.g., Reeve and Sherman19) that they
are not adaptationists in the strict
Darwinian sense but focus instead on
the present fitness consequence of a
trait which cannot logically play any
role in explaining its existence...’’(p.
293).9(Emphasis mine.) Their reason-
ing is that, because present fitness
cannot have caused past evolution, it
cannot play a role in explaining the
present existence of an adaptation.
The views just summarized repre-
sent a sharp break with traditional
evolutionary thinking. For example,
both Mayr and Tinbergen considered
currentfitnessconsequencesto be cen-
tral to a complete understanding of an
adaptation.20–21 Their thinking on this
subject did not entail the absurd as-
sumption that adaptations are caused
by current fitness consequences.
Rather, it rested on the assumption
that continuity of relevant environ-
mental conditions is common, and
hencethatcurrentreproductiveconse-
quences frequently are a guide to what
occurred in the past.
CRITIQUE OF THE EEA CONCEPT
There are several distinct problems
with the concept of the EEA. First,
statementstotheeffectthatourances-
tors were hunter-gatherers for one or
two million years overlooks the fact
that there was considerable evolution
in the hominid line during the Pleis-
tocene. Indeed, the two-million-year
period referred to previously in the
quote by Barkow, Cosmides, and
Tooby2witnessed the appearance and
disappearance of several different
hominid species.10 Paleontologists see
the earlier representatives of this lin-
eage as being very different creatures
with very different ways of life from
modern humans.10 They see modern
Homo sapiens, our species, as having
appeared late in the Pleistocene
(100,000 to 200,000 years ago) and as
having exhibited a wide variety of dis-
tinct hunting and gathering ways of
life. Saying that human beings were
Pleistocene hunter-gatherers for one
or two million years creates a false
picture of stasis during this period.
Second, the statement that 10,000
years is not enough time for evolution-
ary change is hard to defend. Ten
thousandyearswould witness the pass-
ing of 300 to 400 human generations.
In laboratory breeding experiments
with mice, researchers have been able
to obtain nonoverlapping distribu-
tions of behavioral traits in 30 genera-
tions.22 Althoughsuchrapidevolution-
ary change may be rare in nature,
there is evidence of significant evolu-
tionary change over periods of a few
thousand years when conditions are
right.Stanley23 gaveanexampleoffive
new species of cichlids appearing in
LakeNabugaboinfour thousand years
and noted, as well, that major changes
haveoccurredamongpupfishin Death
Valley, California, in a few thousand
years. Zimmerman24 offered evidence
that five species of moths of the genus
Hedylepta have evolved in Hawaii in
only one thousand years. It is true that
all of these changes occurred in crea-
tures with much shorter generation
times than humans. Nevertheless, if
they can evolve into several new spe-
cies in a thousand or so generations,
whycouldnothumans undergo signifi-
cant change in 300 to 400 genera-
tions? In fact, there is strong evidence
that certain physiological adaptations
such as the production of lactase in
adulthoodandthesickle-celltrait have
evolved since the end of the Pleis-
tocene.25 Why could human psychol-
ogynotalsoevolvesome novel traits in
this time span?
There are several distinct
problems with the
concept of the EEA. First,
statements to the effect
that our ancestors were
hunter-gatherers for one
or two million years
overlooks the fact that
there was considerable
evolution in the hominid
line during the
Pleistocene.
ARTICLES
Evolutionary Anthropology 195
Some of the best evidence for the
possibility of rapid evolutionary
change comes from research on Gala-
pagos finches. The Grants and their
coworkers26 have witnessed large
changes in a single generation in re-
sponse to extreme selective pressures
caused first by a drought and then by a
flood. The speed of change seen first-
hand by the Grants stands in sharp
contrast to the expectations of an ear-
liergenerationofevolutionists.Totake
one example, J.B.S. Haldane was con-
vinced that evolutionary change was
too slow to be observed in the field
without extremely fine observations.
In order to make his expectations pre-
cise, he defined a unit for measuring
the rate of evolution. That unit, the
‘‘darwin,’’ was defined as a 1% change
in the length of a metric trait per
million years.27 Examination of the
fossil record led him to believe that
evolution typically occurred at about
this rate. He believed that under artifi-
cialselectiononecouldget much more
rapid change, but that these rapid
rates were rare in nature. As things
have turned out, the Grants’ observa-
tions have produced extremely differ-
ent results.26 The Grants observed
changes of 5 to 6% in a single year
owing to a drought. According to Hal-
dane’s expectations, this should have
required 5 to 6 million years rather
than a single year.28 It is true that
selection reversed its direction after
the drought so that long-term direc-
tional change did not occur. The view
of evolution that comes out of the
Grants’ research is one of continual
rapidchange,butchange that switches
direction frequently. Long periods of
apparent stasis occur often in the fos-
sil record, but this could be because
rapidly alternating directional selec-
tion over long periods can average out
to no change. Thus it would not be
surprising if, in the human line, there
were very little directional change in
either morphological or psychological
traits for 10,000 years. However, this
apparent lack of change is not the
result of insufficient time for evolu-
tion. It is, rather, the result of stabiliz-
ing selection or, perhaps, of wobbling
directional selection that, over long
periods, is the equivalent of stabilizing
selection.
Evolutionary psychologists claim
that human beings are not adapted to
any of the environments that have
occurred since the close of the Pleis-
tocene.Forthistobetrue, there would
have to have been no stabilizing selec-
tion. However, if there were no stabiliz-
ing selection for 300 to 400 genera-
tions, some combination of drift, and
directional and disruptive selection
shouldhavealtered human psychologi-
cal adaptations.
A third serious problem with the
EEA concept as presented by evolu-
tionary psychologists is their insis-
tence that the current fitness conse-
quence of an adaptation can play no
role in explaining its existence. Cos-
mides, Tooby, and Symons are correct
thatpresentfitnesscannot have caused
the origin of an adaptation: Causes
precedeeffectsintime. However,caus-
ing something and playing a role in
explaining it are not necessarily the
same thing. If the form of an adapta-
tion is stable for a very long time and
the environment to which it is ad-
justed is also stable, then measures of
current fitness can play a logical role
in explaining the maintenance of this
adaptation. Such measures have in
fact played such a role in the study of
animal behavior as, for example, in
the Reeve and Sherman19 study criti-
cized by Tooby and Cosmides.9The
changing direction of selection ob-
served by the Grants should caution
us against relying too heavily on a
single instance of such a measure. On
the other hand, numerous measures
may teach us something. Such mea-
sures may provide strong evidence for
stabilizing selection as the main rea-
son for the continuity of adaptations
into the present. Also, in cases where
therelevantenvironmentofan adapta-
tion has not changed over a long pe-
riod, measures of current fitness may
allowindirectinferenceasto the selec-
tive pressures that produced the adap-
tation in the past.
In short, the EEA concept as cur-
rently used by evolutionary psycholo-
gists is unrealistic in terms of what is
now known about human phylogeny,
the pace of evolution, and the role of
stabilizingselectioninmaintaining ad-
aptations. On the other hand, the ex-
pectation that most of human behav-
iorinmodern environments is adaptive
presents problems. The most fre-
quently cited problem is that individu-
als with greater resources tend not to
use these resources for greater repro-
duction.29–30 Vasectomies, voluntary
childlessness,andextensiveuse of con-
traception to lower lifetime fertility
below replacement are good candi-
dates for behavior that natural selec-
tion does not favor. It is, in my view, a
reasonable hypothesis that some of
modernurbanhumanbehavioris mal-
adaptive because of recent changes in
human environments. However, much
of the relevant environmental change
is of more recent origin than the end
of the Pleistocene and does not appear
tobecloselytiedtothe transition from
foraging to food production. Modern
contraceptive technology has only a
short history. Evidence further indi-
cates that modern low rates of fertility
are also historically novel as a wide-
spread phenomenon. The more tradi-
tional societies that anthropologists
study characteristically have total fer-
tility rates in the range of five to eight
births. (The total fertility rate is a
population-wide average: the average
number of children a women has if
shelivestotheend of her fertile years.)
Modern populations that make exten-
sive use of contraception have total
fertilityratesthatcluster around two.31
Ethnographic analogy suggests that
the former higher fertility rates were
characteristic of most of post-Pleis-
tocene human history. It is also rel-
evant that, given typical death rates
before modern medicine, populations
with total fertility rates of three or less
wouldprobablyhavegone extinct. This
sort of evidence suggests that the de-
Evolutionary
psychologists claim that
human beings are not
adapted to any of the
environments that have
occurred since the close
of the Pleistocene. For
this to be true, there
would have to have
been no stabilizing
selection.
196 Evolutionary Anthropology
ARTICLES
mography of modern societies is quite
different from that of most post-
Pleistocene societies. We need a con-
cept that can deal realistically with
this possibility.
EventhoughIhave judged the evolu-
tionary psychologists harshly for their
relianceontheEEA,Ithink their work
has great value and, in fact, provides a
keyconceptthatwillallowbetter char-
acterization of the relationship be-
tween adaptations and environments.
This key, in my opinion, is their view
that the human mind (and, presum-
ably, the minds of other animals) con-
tains a large number of special-pur-
pose decision-making mechanisms.
They draw on many sources to defend
this concept, which, unlike the con-
ceptoftheEEA,Iconsidertobeuseful
and basically accurate.
It may be worth noting at this point
that the strength of the evolutionary
psychologists falls in an area that they
observe directly and closely in their
empirical research. The difficulties
with the EEA concept have to do pri-
marily with paleontology and field ob-
servationsofevolutionarychange, sub-
jects that they do not study at all in
their empirical research. That they
shouldberightabout things they study
first-hand,butwrongabout things they
knowonlysecond-hand,isnot surpris-
ing.
MANY SPECIAL-PURPOSE
MECHANISMS
The idea that the human mind con-
tains many special-purpose decision-
making mechanisms derives from
studies that indicate that human deci-
sion making differs with the specific
problem being addressed. Most deci-
sion making is, as the evolutionary
psychologists say, domain-specific or
content-specific. For example, experi-
mental results indicate that people
find symmetrical faces more attractive
thanasymmetricalones.Further,there
isevidencethatpeoplewith more sym-
metrical faces are healthier and more
parasite-resistant. There is similar evi-
dence that faces having close to aver-
age proportions are more attractive
and, again, are an accurate signal of
health and parasite resistance.32 One
problem that people faced in ancestral
environmentswasthat of mate choice.
Healthiermates,especiallythoseresis-
tant to local parasites, are better than
less healthy ones. So natural selection
seems to have favored a propensity to
find potential mates with symmetrical
and average faces more attractive. It
did not, however, favor an awareness
that such people are healthier. Nor did
it favor a conscious goal of high repro-
ductivesuccessandadesiretoidentify
healthy mates in order to accomplish
this goal. Selection merely favored a
perception that symmetrical faces and
average faces are more attractive, thus
motivating individuals to seek mates
with these characteristics.
When we turn to other types of
decisions, such as, food choices, for
example, we find a different set of
decision-making criteria. Evolution-
ary psychologists describe this as a
differentdecision-makingalgorithm.33
Thedietofthe Pleistocene, it is hypoth-
esized, contained a large amount of
fibrous tubers and other such veg-
etables. These foods contained low
levels of nutrition and high levels of
toxins. Ripe fruit, in contrast, con-
tained high levels of nutrition (sugars)
andlowlevels of toxins. However, fruit
was scarcer than fibrous vegetables or
unripe fruit. Thus, natural selection
favored a preference for ripe fruit over
other foods. This preference came in
the form of a preference for things
with a sweet taste. For similar rea-
sons, people evolved a preference for
salty flavors and the flavor of animal
fats.Ontheotherhand,fiber, although
necessary for healthy function of the
digestive system, tended to occur in
the diet in greater abundance than
was necessary. Thus no preference for
fiber evolved. It is hypothesized that in
ancestral environments these prefer-
ences motivated people to come as
close as their circumstances allowed
to optimal diets.
However in modern environments,
the abundance of different types of
foods is vastly different, and these
preferences often motivate people to
choosedietsthatare much less healthy
than are possible in their circum-
stances.33 Selection building on the
variation available did not favor an
intellectual understanding of the sort
found in a dietitian’s handbook of
what constitutes a healthy diet plus a
conscious desire to approximate such
a diet as closely as possible. If it had,
the mechanism of food choice would
be less domain-specific and would
work better in modern environments.
However,such a more domain-general
mechanismcouldnotevolve,giventhe
natureofour prehuman ancestors and
thetypesofvariation available to selec-
tion in ancestral populations.
There is evidence for domain-spe-
cific mechanisms dealing with a num-
ber of other mental tasks, such as
facial recognition, the acquisition of
language, maternal bonding, inbreed-
ing avoidance, and the detection of
cheaters in social exchange.2,34 Evolu-
tionary psychologists expect eventu-
ally to uncover evidence for a large
number of specialized mental mod-
ules. There are some more general-
purposemechanismsaswell,butnone
are as general as a single, all-purpose
fitness maximizer.
Thehypothesisthatthe human mind
(as well as the minds of other animals)
consists of a large number of special-
ized tools has been reinforced by the
evidencefromneuroscience that differ-
ent mental tasks are processed by very
different neural networks.34 This has
led to the working hypothesis that
differentdecision-makingmodulesare
contained in different neural net-
works, just as the different algorithms
of computer software are contained in
different lines of computer program-
ming instructions.9Damage to one set
ofneuralnetworks causes specific algo-
rithms to fail, but not others con-
tainedinundamagedneuralnetworks.
This situation occurs, for example,
when damage to one area of the brain
disrupts certain mental functions but
not others.
The idea that the human
mind contains many
special-purpose
decision-making
mechanisms derives
from studies that indicate
that human decision
making differs with the
specific problem being
addressed.
ARTICLES
Evolutionary Anthropology 197
Evolutionary psychologists see do-
main specificity as logically tied to
their belief that fitness consequences
outside the EEA are not relevant. They
argue that those who believe in adap-
tiveness outside the EEA must believe
that the human mind consists of only
oneorafewgeneral-purposedecision-
making mechanisms. As I have noted,
a more general mechanism for food
choice, such as a scientific dietitian’s
understanding of human nutritional
needs,wouldbemore effective in novel
environments. General mechanisms,
it seems, are better at dealing with
novelty. The ultimate general-fitness-
maximizing mechanism would con-
sistofadesireto maximize one’sinclu-
sive fitness, accompanied by a desire
tounderstandone’senvironment suffi-
ciently to do this. Such a mechanism
would operate effectively in any novel
environment, but only if the organism
were able to gain sufficient knowledge
of its environment before dying of
starvation or succumbing to some
other hazard. Evolutionary psycholo-
gists are convinced that expecting ad-
aptationstoenhance reproduction out-
side the EEA must entail some such
unrealistic assumption of a general-
ized mechanism.
While I accept the evolutionary psy-
chologists’ hypothesis of multiple spe-
cific mechanisms, I do not accept their
criticisms of human evolutionary
ecologists who choose not to incorpo-
rate this assumption into their re-
search.8,35,36 Evolutionary psycholo-
gists claim that the work of these
researchers is completely miscon-
ceived because it fails to incorporate
this hypothesis. They further claim
that evolutionary ecologists assume
implicitly the presence of one or a few
general-fitness-maximizing mecha-
nisms.2However, as I see it, these
criticisms misread the work of evolu-
tionary ecologists. These researchers
conceivedoftheir work, in Tinbergen’s
and Mayr’s terms, as a matter of con-
structingandtestingfunctionalor ulti-
mate cause models.20,21 For these re-
searchers, how the human mind is
structured was a matter of proximate-
cause analysis; they were doing an
ultimate-causeanalysis.Whileaproxi-
mate-cause analysis would be comple-
mentary to their work, it need not be a
part of it. They made no assumptions
about the structure of the human
mind. They were studying the way
behavior maps onto different environ-
ments, not the structure of the mind.
Most of these researchers would agree
with Tinbergen and Mayr that a com-
plete understanding of an adaptation
wouldrequirebothultimate and proxi-
mate analyses. Their pursuit only of
ultimate analyses was simply a good
place to start. They expected that fu-
ture research would eventually turn to
proximate-level analyses.37
ADAPTIVELY RELEVANT
ENVIRONMENTS
The adaptively relevant environ-
ment (ARE) of an evolved adaptation
consists of those features of the envi-
ronment that the mechanism must
interactwithinorder to confer a repro-
ductive advantage. As a rule, an adap-
tation needs to interact with only a
fewselectedelementsoutof the organ-
ism’s total environment in order to
confer its reproductive advantage. Dif-
ferent adaptations interact with very
different features of the environment.
Thus environmental novelties have
quite specific effects, disrupting some
adaptationsbutnotothers. When long-
term changes occur in the environ-
ment of a population and the popula-
tion survives in the new environment,
those adaptations having changed
AREs undergo evolution. Adaptations
having unaltered AREs remain the
same. Evolution is mosaic.
Asnotedearlier,the concept of AREs
is a logical complement to the idea
that the human mind consists of many
specific-purpose information-process-
ing mechanisms rather than a few
general-purpose ones.38 Each of these
specificinformation-processingmecha-
nisms is an evolved adaptation. Each
was designed by natural selection to
solve a specific set of problems that
ancestors needed to solve in order to
survive and reproduce. For an adapta-
tion to solve a specific problem, it
must interact in a consistent way with
specific features of the environment.
Whether an adaptation can be ex-
pected to produce an adaptive effect is
an empirical question. To answer it we
need to know what its ARE is and
whether it is intact. More important,
the best way to elucidate the design
features of adaptations is to study
them in both their relevant environ-
ments and novel environments.39
Studying adaptations in both situa-
tionswillrevealexactlywhichfeatures
of the environment an adaptation is
connected to and exactly how it is
designed to interact with these envi-
ronmental features. The anthropologi-
cal study of more traditional societies
can be crucial in this sort of empirical
work. Many adaptations can be ob-
served interacting with their relevant
environments in these societies, but
not in modern urban societies. It
should be emphasized that opportuni-
ties of this sort are not limited to
foraging societies. The relevant envi-
ronment of many adaptations does
not include a foraging economy, as
will be seen in the discussion of spe-
cific AREs to follow.
The basic idea conveyed by the con-
cept of the EEA is that adaptations
evolved in past environments and can-
not necessarily be expected to func-
tion adaptively in novel ones. This
idea is correct. The concept of AREs is
meanttoincorporatethisinsight with-
out making the unrealistic assump-
tions associated with the EEA concept
I noted earlier.
Although the ARE concept is tied to
the idea that the human mind consists
ofmanyspecificpsychologicaladapta-
tions, it is also compatible with the
view that the human mind consists of
ahierarchyof mechanisms, with some
being more general in purpose than
others.40–41
The exact nature of this concept can
best be made clear with a set of ex-
amples. The examples I will present
aremeant to clarify the exact nature of
theconcept,suggestwaysinwhich the
concept can guide empirical research,
and argue that the best way to under-
stand a proximate mechanism is to
study its interaction with its ARE and
with novel environments. The ex-
For an adaptation to
solve a specific problem,
it must interact in a
consistent way with
specific features of the
environment.
198 Evolutionary Anthropology
ARTICLES
amples used are those of status striv-
ing,maternalbonding,andthe Wester-
marck effect.
These examples are based on theo-
retical models currently found in the
literature of evolutionary psychology,
evolutionaryecology,andrelated disci-
plines. All of these models are contro-
versial to some extent, as are almost
all evolutionary models of human psy-
chology and behavior. I will not at-
tempttodealwithallofthetheoretical
difficulties raised by each model or to
assessthoroughlyalltheevidencecon-
firmingorinvalidating the accuracy or
usefulness of each model. Because the
purpose of the discussion is to illus-
trate the ARE concept, I simply pre-
sent the models more or less as they
appear in the literature and then con-
struct an accompanying model of the
ARE of each adaptation. Next I review
briefly some empirical research to
illustrate ways that the ARE concept
can guide research. Future research
may lead to the modification or even
abandonment of many of the specific
models discussed. Nevertheless, I be-
lieve these models of evolved psycho-
logical mechanisms are viable guides
toresearchatpresent.Ifurtherbelieve
that the expansion of these models to
include their relevant environments
makes them more precise guides to
research.
THE ADAPTIVELY RELEVANT
ENVIRONMENT OF STATUS
STRIVING
Human beings have a propensity to
identify the criteria that define social
status in their culture and to use this
information to raise their own status
when it is realistic to do so. It has been
suggested as a hypothesis that evolved
psychologicaladaptationsunderlie this
propensity.13,42–44 These adaptations
consist, first of all, of a propensity to
identify those things in one’s immedi-
ate social environment that constitute
resources for reproduction. These
things vary from one society to an-
other. Among Yanomamo males,
fierceness (the ability to threaten and
use violence to one’s advantage) is the
primary criterion of status. Among
Ache males, it is hunting reputation.
In many other groups, it is wealth.
Once people have identified these
things, they seek them for themselves
and evaluate the status of others ac-
cording to their success in acquiring
these things. They also socialize their
children to seek these things.45 This
hypothesis also states that people are
realistic in identifying as standards of
successthingsthat really do constitute
resources for reproduction in their
environment, and that, as their envi-
ronment changes, they modify their
standards to keep them realistic.
This hypothesis can be falsified by
showing a lack of correlation between
culturallydefinedstatusand reproduc-
tive success and can be confirmed, to a
degree, by showing the presence of such
a correlation in many societies. The
hypothesis has been tested in many
settings with mixed results.13,29,42–45
Some researchers take these mixed
results to mean that the status-striving
propensities of human beings are not
expressions of evolved psychological
adaptation.46 They see these results as
supporting the widely accepted view
that cultural beliefs and values vary
randomly with respect to what is bio-
logically adaptive.47 This view, applied
to cultural criteria of status, suggests
that these standards sometimes corre-
spond to resources for reproductive
success, but just as often encourage
people to do things that diminish re-
productivesuccess.Mixedresults seem
to fit this expectation. Others take
these mixed results as proof that it is
meaningless to test this hypothesis
outside foraging societies because
other types of societies are too far
removed from the EEA.8,18 Still others
have suggested that the results as a
whole make a kind of evolutionary
sense because the hypothesis is gener-
ally confirmed in the more traditional
societies, but not in urban, industrial
societies, which represent evolutionarily
more novel environments.13,45
Thinking in terms of AREs suggests
that the latter view can be made more
precise and testable by identifying the
ARE of status striving. Both logic and
data suggest that the ARE of status
striving includes polygyny and limited
contraception. By limited contracep-
tion, I mean not making extensive use
of any practice that intentionally re-
duces the number of children parents
rear.Intraditionalsocieties, such prac-
Figure 1. A high-status Yomut Turkmen man (in Northern Iran in 1967) with part of his family. The
two adult women to the left are his two wives. The woman closest to him is a daughter-in-law
holding a grandson. The rest of those shown are seven of his fourteen children.
Both logic and data
suggest that the ARE of
status striving includes
polygyny and limited
contraception.
ARTICLES
Evolutionary Anthropology 199
tices can include postpartum taboos
and infanticide (which, although not a
contraceptive, can have a very similar
function).Intraditional societies, these
practices are common, but they rarely
reduce the total fertility rate below
four. The average total fertility rate for
such societies is around six.31 If the
total fertility rate does fall below four
in a traditional society, it is usually the
resultofahighfrequency of pathologi-
cal sterility. Societies that fall in this
category are often called natural fertil-
ity societies. In modern societies, in
contrast,theuseofmodern contracep-
tives often reduces the total fertility
rate to something near to, or even
below, two. I refer to this latter situa-
tion as extensive contraception. It is a
reasonablehypothesisthat most of the
social environments in which humans
evolved were characterized by po-
lygyny (defined in terms of breeding
structure, not marriage) and limited
contraception. This means that, dur-
ing human evolution, a propensity to
strive for status would have consis-
tently interacted with these features.
A meaningful pattern emerges if we
sort the various tests of the cultural
success hypothesis according to the
types of society they derive from. Soci-
eties fall into three categories in terms
of the features defining the ARE of
status-striving.Thesearesocieties with
polygyny and limited contraception;
with socially-imposed monogamy and
limited contraception; and with so-
cially-imposed monogamy and exten-
sive contraception. The first category,
polygyny and limited contraception,
represents societies in which the ARE
of status striving is intact. A correla-
tionbetweenreproductivesuccessand
status for men has been reported for
many societies falling into this cat-
egory,includingthe YomutTurkmen,42,43
the Yanomamo,48,49 the Ache,50 nine-
teenth-century Mormons,51 the Kipsi-
gis,52 the Mukogodo,53 the Bakkarwal,54
the Gabbra,55 the Havasupai,56 the Do-
gon (B.I. Strassmann, personal com-
munication), and Thai villagers.57 In
all eleven of these societies, there is a
statisticallysignificantpositive correla-
tion between status and reproductive
success. For three of these societies, a
statistically positive correlation be-
tween status and reproductive success
was also reported for women. For the
others, no data has been reported for
women. These results are consistent
with the idea that polygyny and lim-
ited contraception constitute the ARE,
orrelevantenvironment, of status striv-
ing.
At the opposite end of the spectrum
are modern societies in which socially
imposed monogamy and extensive
contraception are the rule. In these
societies, there is usually either an
inverse correlation or no correlation
between status and reproductive suc-
cess.45,58 In between are a large num-
ber of historically recent and contem-
porary societies characterized by
socially imposed monogamy but lim-
ited contraception. In the majority of
thesesocietiesthereisa positive corre-
lation between reproductive success
and status, although the pattern is less
pronounced. This makes sense, given
that, in these societies, the ARE of
status striving has been partially dis-
rupted.
The demographic studies I have re-
viewed support the proposition that
the psychological mechanisms that
give rise to human status striving
evolved in environments character-
ized by polygyny and limited contra-
ception, and that these two features
are central to the ARE of status striv-
ing. The hypothetical ARE model I am
suggesting assumes that polygyny and
limited contraception were very close
to being universal characteristics of
early human environments. Thus,
people did not evolve a propensity to
track the environment to determine
whether such features were present.
Suchtrackingwouldhave been a waste
of resources. It was more efficient
simply to identify the criteria for sta-
tus and strive to improve one’s status
when the costs were not too high.
Thus, the desire to strive for status
occurs in many recent and contempo-
rary environments despite limitations
on polygyny and despite the presence
of extensive contraception. Because
polygyny is defined in terms of breed-
ing structure, not marriage rules, it is
better to speak of more limited po-
lygyny rather than the absence of po-
lygyny. Through serial marriage and
clandestine unions, high-status men
probably still enjoy higher access to
mates than do low-status males. The
varianceinaccesstomatesisprobably
still higher among males than among
females in modern settings. However,
contraception limits the extent to
which the extra access to mates
through clandestine unions is trans-
lated into reproductive success.45 Ex-
tensive contraception, more than limi-
tation of polygyny, is probably the
important novel environmental ele-
ment disrupting the ARE of status
striving.
Long after the close of the Pleis-
tocene, the ARE of status striving was
the most common condition experi-
enced by human beings. This ARE
consists of a social environment char-
acterizedbypolygynyandlimitedcon-
traception. Thus, for the most part,
stabilizing selection has maintained
the psychological adaptations that un-
derliestatusstriving. Even today,much
of the world’s human population still
lives in such an environment. How-
ever, the conditions familiar to most
readers of this paper are the recent
novel conditions of socially-imposed
monogamy and extensive contracep-
tion.Itisimportantforus to recognize
that our first-hand experience are lim-
ited to evolutionary novel conditions.
The preceding discussion does not
address the question of why socially-
imposed monogamy or extensive con-
traception have come to characterize
modern experience. The value of the
concept of AREs lies partly in focusing
our attention on questions for future
research. In this case, these questions
include the reasons for the historic
development of socially-imposed mo-
nogamy and extensive contraception.
For discussions of the origins of so-
cially-imposed monogamy, see Betzig59
and McDonald.60 For discussion of the
motivationsunderlyingtheextensiveuse
ofcontraceptioninmodernsocieties,see
The hypothetical ARE
model I am suggesting
assumes that polygyny
and limited
contraception were very
close to being universal
characteristics of early
human environments.
200 Evolutionary Anthropology
ARTICLES
Kaplan and coworkers,58 Irons,61 and
Turke.62
THE ADAPTIVELY RELEVANT
ENVIRONMENT
OF MATERNAL BONDING
The model of maternal bonding I
use here is derived from the work of
Daly and Wilson (pp. 69–72).6They
suggested that human maternal bond-
ing consists of three processes that
occur following birth. The first con-
sists of the new mother’s assessing her
resources for child rearing and the
quality, in terms of survival and even-
tual reproductive chances, of her new
infant. This stage is often accompa-
nied by emotional indifference toward
the infant, which often alarms the
staff of modern hospitals, especially
those that attempt to encourage bond-
ing through immediate and continual
contact between mother and infant in
the days immediately after birth. The
mother is attentive to the physical
condition of the infant and to its re-
sponsiveness. She also assesses her
resources for child rearing: Does she
have a supportive husband or has he
abandoned her? Is she in good health?
Is she surrounded by supportive kin
and associates or is she alone and
barely able to care for herself? Does
she already have older children who
demandalltheresourcesshe has, leav-
ing none for the new child? In many of
the environments of human evolution,
this period of assessment could lead to
infant abandonment, or infanticide. It
is important to note that in many
traditionalsocieties,womengave birth
in a secluded location with only a few
female kin present. The decision as to
whether or not to raise the child could
then proceed without the interference
of men or unrelated women. In these
environments, the assessment process
led mothers to abandon infants when
the chances of successful child-rear-
ing were too low to warrant the costs.
This could occur either because of the
condition of the child or because of
the condition and resources of the
mother.
The second process of maternal
bonding occurs over the course of
about a week after birth and consists
of the establishment of an individual-
ized love. During the first few days
after birth, new mothers become
acutely aware of the characteristics of
their new infants, their voices, distinc-
tive features, and smells. During this
time they develop a feeling that their
newinfantisuniquelywonderful.This
process assures that maternal nurtur-
ing will not be parasitized by unre-
latedinfants,andhas parallels in many
nonhuman species. The third process,
which occurs over many years, con-
sists of a gradual deepening of mater-
nal love. Here the mother’s love (cold
as it may sound) tracks the rising
reproductive value of the child.
These processes develop out of the
interaction of the experience of birth
and the evolved psychological mecha-
nisms of the mother. In the environ-
ments of evolution, this three-stage
process, in effect, made it possible for
mothers to assess accurately the inclu-
sive-fitness costs and benefits of vari-
ous forms of parental care. It is impor-
tanttonotethat this set of mechanisms
probably has a history (with some
modification) going back as far as the
origin of the mammalian line. All fe-
male mammals make large commit-
ments to maternal care and need to
assess when infants are worth caring
for and when they should be aban-
doned. The process of assessing the
costs and benefits of maternal care
may in fact go back further among
premammalian ancestors who made
large commitments to parental care.
The concept of AREs is not tied to a
particular geological period such as
the Pleistocene. Some adaptations
have a much longer history than oth-
Figure 2. A strongly bonded Yomut Turkmen mother and son in Northern Iran about 1966. All female mammals
make large
commitments to
maternal care and need
to assess when infants
are worth caring for and
when they should be
abandoned.
ARTICLES
Evolutionary Anthropology 201
ers. One problematic feature of the
EEA concept is that it encourages a
sort of Pleistocentrism in which all
human psychological adaptations are
tightly tied to the conditions of Pleis-
tocene foraging societies. Probably
many adaptations, like the mecha-
nisms underlying maternal bonding,
have continued to produce their adap-
tive effects through several geological
periods (and chronospecies), while
many conditions not relevant to their
functioning have changed. This again
relates to the fact that evolution is
mosaic.Thelonghistory of some adap-
tations also suggests that the concept
of phylogenetic heritage may be a use-
ful companion to the ARE concept.10
The ARE that is relevant to these
processes consists of an environment
in which females are free to make
decisions concerning these matters on
their own with the support of close kin
and without the interference of unre-
lated individuals. The advent of hospi-
tal births has disrupted the interaction
of these psychological mechanisms
andtheirenvironments.Thecontinual
contact between mother and infant
that occurred in the environments of
evolution was disrupted by hospital
procedures, weakening the bonding
that occurred in the past. Research
reviewed by Daly and Wilson (pp. 69–
72)6suggested that these procedures
ledtomanyundesirable consequences.
Mothers denied early contact breast
fed their infants for shorter periods
and were more likely to abuse them.
Research led to claims and counter-
claims about the nature of bonding
and the ideal hospital procedures.
However, Daly and Wilson point out
that the results of the large body of
research now in print are consistent
with their model.
Again the evidence indicates that
the processes of maternal bonding are
evolved adaptations that have to inter-
act with specific environmental fea-
tures for them to confer the benefit of
optimal apportionment of maternal
care, the central feature being control
of mother-infant interaction by the
mother herself rather than the unre-
lated individuals that make up hospi-
tal staff. In all probability, this set of
psychological adaptations operated as
it did in the environments of evolution
long after the close of the Pleistocene
and the passing of a foraging way of
life. Eventually, however, for a portion
of the world’s population, a novel envi-
ronment did replace the relevant envi-
ronment and the mechanisms ceased
to achieve their former effect. The
significant transition here was the re-
placement of a personal kin network
with an impersonal bureaucracy as
the social environment of childbirth.
This is something quite different from
the transition from foraging to food
production. Historically, this is quite
recent and is associated with hospital
births.
Discussion of maternal bonding
highlightstheprobabilitythatin much
ofhumanevolutionmanyinfants were
abandonedorkilledshortlyafter birth.
There is an important moral issue
here. The possibility that hospital pro-
cedures can raise the risk of child
abuse is another serious moral issue. I
do not address these issues here. Suf-
fice it to say that grappling with these
difficult moral concerns will, in all
probability, produce better ethical de-
cisions if the relevant psychological
adaptations and their AREs are better
understood.
THE ADAPTIVELY RELEVANT
ENVIRONMENT OF THE
WESTERMARCK EFFECT
My model of the Westermarck effect
is taken primarily from the research of
Arthur Wolf, who, with his co-worker
Huang, has produced an excellent and
thorough study of this psychological
adaptation.63,64 Briefly, according to
the hypothesized model, the Wester-
marck effect occurs when two indi-
viduals live in intimate contact while
one of them is in the age range of
roughly newborn to five years. This
psychological mechanism eliminates
sexualattractionbetweenthe individu-
als in question later, when they are
both of reproductive age. It thus dis-
courages close inbreeding of the form
that is defined as incest by most cul-
tures. There are parallel adaptations
in many animal species.65 Again, in all
probability, we are dealing with an
adaptationthat existed long before the
Pleistocene. It is well documented that
close inbreeding produces lowered vi-
ability of offspring, a condition known
asinbreedingdepression.63–65 TheWes-
termarck effect, which probably ex-
isted in prehuman ancestors, lowered
the risk of inbreeding depression.
There are other mechanisms that join
forces with the Westermarck effect to
prevent inbreeding depression.63,64
However, for the current discussion I
will focus on just one mechanism, the
Westermarck effect, and will concen-
trate on Wolf and Huang’s study of it.
Thisstudydealswith the consequences
of the practice of ‘‘minor marriages’’ in
Taiwan.Minormarriagesarearranged,
in effect, when mothers give their
daughters, shortly after birth, to other
families who raise them to be wives
for the sons of their households. Here
individuals intended for marriage are
raised in intimate contact, with the
result that the marriage is usually not
highly successful. Wolf and Huang
show that such marriages have 31%
lower fertility and a threefold higher
divorce rate than a control sample of
marriages between individuals not
reared in intimate contact. Wolf and
Huang examined a large sample of
minor and major marriages and did
elaborate analyses to eliminate alter-
nate explanations of the lower fertility
and higher divorce rate of minor mar-
riages. These alternate explanations
include the effects of adoption itself,
the poorer health of adopted daugh-
ters-in-law, poverty, the lesser prestige
of such marriages, and the effects of
age at marriage and sibling rivalry.
Theirdataanalysesclearlysupportthe
hypotheses that there is a Wester-
marck effect, and that this mechanism
produces a nonadaptive result when it
. . . the evidence
indicates that the
processes of maternal
bonding are evolved
adaptations that have to
interact with specific
environmental features
for them to confer the
benefit of optimal
apportionment of
maternal care . . .
202 Evolutionary Anthropology
ARTICLES
encounterstheevolutionarilynovelin-
stitution of minor marriage.
There is evidence for the Wester-
marck effect in some other studies as
well. These are reviewed thoroughly in
Wolf’s1995bookSexualAttractionand
Childhood Association.63 In my opin-
ion, he makes a good case that the
available data support the existence of
the Westermarck effect. Certainly the
evidence is good enough to justify
further research. The literature on the
Westermarck effect already spells out
clearly what the adaptation’s ARE is.
This is the case because most of the
empirical studies are of cases in which
the adaptation’s relevant environment
has been disrupted and nonadaptive
behaviors have emerged.
The ARE of the Westermarck effect
is a social environment in which close
kin, siblings, parents, and children are
in intimate contact during the critical
period of the first two or three years of
a child’s life, and in which intimate
contact is rare between nonkin or dis-
tant kin when one or both parties are
in the critical age range of newborn to
three years. Given what we know of
more traditional societies, it is prob-
able that this condition occurred in
most evolving human populations and
probably did so as well for a very long
time among our prehuman ancestors.
The fact that it works most of the time
but not always is probably typical of
many adaptations. This fact is also
probably relevant to the fact that the
Westermarck effect works in conjunc-
tion with other mechanisms evolved
to prevent inbreeding depression.63
A lot has been written on the subject
of incest avoidance in human beings.
One piece of data that has received a
great deal of attention in this litera-
ture is the fact that in a small number
of highly stratified societies the royal
heirs at the top of the social hierarchy
are required to marry their siblings.66
There is also evidence that in the case
of the Ptolemies, who practice sibling
marriage for royal heirs, the practice
extended to the middle class. Despite
these well-known cases, the practice is
very rare. Nevertheless, there has been
much speculation as to why it should
occur at all in the face of the Wester-
marck effect.
At least one possible explanation is
that among the rare groups that prac-
tice sibling marriage, rearing prac-
tices prevent intimate contact be-
tween siblings during the critical
period.63,64–66 It is common practice
for the heir apparent at the top of a
highly stratified society to be reared
separately from siblings. This could
remove one barrier to sibling mar-
riage.Other forces would then have an
easier time leading to sibling mar-
riage. This is a question that could be
the focus of future research. It is also
possible that, like minor marriage in
Taiwan, these practices did run afoul
of the Westermarck effect, but that the
cost of such difficulty was offset by
some advantages. Wolf hypothesizes
that the Taiwanese preference for mi-
nor marriages was the result of a
male-biased sex ratio and severe com-
petition for wives. Thus, while the
practice had its cost, it also had its
compensating benefits. The sibling
marriages of the middle class of Ptole-
maic Egypt may also have had some
benefit as well as cost. This also can be
a question for future research. One
advantageoftheAREconceptisthatit
does not presume that adaptations
always produce adaptive effects, but
rather constructs testable models of
the conditions under which they will
and will not produce such effects.
WHY ADAPTIVELY RELEVANT
ENVIRONMENTS INSTEAD
OF THE ENVIRONMENT OF
EVOLUTIONARY ADAPTEDNESS?
The last few centuries of human
history have been characterized by
rapid and ever-accelerating cultural
change. The result is that many as-
pects of modern urban environments
are evolutionarily novel. Given this, it
should not be surprising that much of
human behavior in these environ-
ments is not adaptive. Modern low
fertility, extensive use of contracep-
tion, and voluntary childlessness
wouldseemtobeprimecandidatesfor
nonadaptive behaviors. The use of ad-
dictive drugs, modern eating habits,
and the sedentary nature of urban life
are other candidates for nonadaptive
behaviors. This is the world that most
of the readers of this article know
first-hand. It naturally has made a
deep impression on those who study
human behavior and psychology in
evolutionary terms. The EEA concept
has no doubt been useful in reminding
people that evolutionary theory does
notpredict that everything is adaptive,
as well as that the apparent nonadap-
tiveness of much of modern behavior
should not deter us from exploring the
theory of evolution as a source of
insight into human behavior, thought,
and feelings. I believe the concept
derives much of its appeal from this.
However, as I see it, the EEA con-
cepthastoomuchbaggagethatcan no
longer be defended. We need a new
concept. I suggest the ARE as this new
concept. It is, I believe, more consis-
tent, among other things, with the
view that the human mind consists of
many specialized adaptations. The
concept is also consistent with the fact
that evolution can occur rapidly and is
always mosaic, and that traits are un-
likely to persist unchanged in the ab-
sence of stabilizing selection. Further-
more,itisconsistent with the extensive
evidencefortheadaptiveness of behav-
ior in more traditional societies. The
main advantage of the concept, how-
ever, as I see it, is that it can focus
empirical research on the interaction
of specific adaptations and their rel-
evant environments. We cannot fully
understand the design of adaptations
without examining their interaction
with a range of specific environmental
conditions.
ACKNOWLEDGMENTS
Benjamin Campbell, John Fleagle,
Kristen Hawkes, Raymond Hames,
SarahHrdy,WilliamLukas, Paul Quin-
dry, Eric Smith, and Mark Turner read
earlier drafts of this paper and made a
number of very helpful suggestions. I
would also like to note that the ARE
concept first occurred to me while
reading Daniel Pe´russe’s 1993 BBS
article.45 Although Pe´russe does not
spell out such a concept specifically in
that paper, he comes very close to
. . . the EEA concept has
too much baggage that
can no longer be
defended. We need a
new concept.
ARTICLES
Evolutionary Anthropology 203
doing so, and this is what inspired me
todeveloptheconceptspecifically.Any
flaws in this paper are of course my
responsibility. I thank R.L. Wadley, M.
Pouliadakis, and B.L. Strassman for
permission to cite unpublished data.
REFERENCES
1Alexander RD (1979) Darwinism and Human
Affairs. Seattle, WA: University of Washington
Press.
2Barkow JH, Cosmides L, Tooby J (eds) (1992)
The Adapted Mind; Evolutionary Psychology and
the Generation of Culture. Oxford: Oxford Univer-
sity Press.
3Betzig L, Borgerhoff Mulder M, Turke P (eds)
(1988)Human ReproductiveBehaviour: ADarwin-
ian Perspective. Cambridge: Cambridge Univer-
sity Press.
4ChagnonNA, Irons W (eds) (1979) Evolutionary
Biologyand HumanSocial Behavior: An Anthropo-
logical Perspective. North Scituate, MA: Duxbury
Press.
5Daly M, Wilson M (1983) Sex, Evolution, and
Behavior. 2nd Edition. Boston, MA: PWS Publishers.
6Daly M, Wilson M (1988) Homicide. New York:
Aldine de Gruyter.
7Smith EA, Winterhalder B (eds) (1992) Evolu-
tionary Ecology and Human Behavior. New York:
Aldine de Gruyter.
8Symons D (1992) On the use and misuse of
Darwinism in the study of human behavior. In
BarkowJH, CsomidesL, ToobyJ (eds)The Adapted
Mind: Evolutionary Psychology and the Genera-
tion of Culture, pp 137–162. Oxford: Oxford Uni-
versity Press.
9Tooby J, Cosmides L (1997) Think again. In
Betzig L (ed) Human Nature: A Critical Reader,
pp. 292–294. Oxford: Oxford University Press.
10 Foley R (1996) The adaptive legacy of human
evolution: A search for the environment of evolu-
tionary adaptedness. Evol Anthropol 4:194–203.
11 Alexander RD (1990) Epigenetic rules and
Darwinian algorithms. Ethol and Sociobiol 11:
241–303.
12 Betzig L (1998) Not whether to count babies,
but which. In Crawford C, Krebs D (eds) Hand-
book of Evolutionary Psychology: Ideas, Is-
sues, and Applications, pp. 265–273. London:
Lawrence Erlbaum.
13 Irons W (1990) Let’s make our perspectives
broader rather narrower: A comment on Turke’s
‘‘Which humans behave adaptively, and why does
it matter?’’ and on the so-called DA-DP debate.
Ethol and Sociobiol 11:361–374.
14 TurkePW(1990) Which humansbehave adap-
tively, and why does it matter? Ethol and Socio-
biol 11:339.
15 Wright R (1994) The Moral Animal: Evolution-
ary Psychology and Everyday Life. New York:
Vintage Books.
16 Bowlby J (1969) Attachment and Loss, Vol. I:
Attachment. New York: Basic Books.
17 Bowlby J (1973) Attachment and Loss, Vol. II:
Separation, Anxiety, and Anger. New York: Basic
Books.
18 Symons D (1979) The Evolution of Human
Sexuality. Oxford: Oxford University Press.
19 Reeve HK, Sherman PW (1993) Adaptation
and the goals of evolutionary research. Quarterly
Review of Biology 68:1–32.
20 Mayr E (1961) Cause and effect in biology. In
Mayr E (ed) Evolution and the Diversity of Life,
pp. 359–371. Cambridge, MA: Harvard University
Press.
21 Tinbergen N (1963) On aims and methods of
ethology. Zeitschrift fu¨r Tierpsychologie 20:410–
433.
22 DeFries JC (1980) Genetics of animal and
human behavior. In Barlow GW, Silverberg J
(eds) Sociobiology: Beyond Nature/Nurture? pp
273–294. AAAS Symposium 35. Boulder, CO:
Westview Press.
23 StanleySM (1981)The NewEvolutionary Time-
table: Fossils, Genes, and the Origin of Species.
New York: Basic Books.
24 Zimmerman EC (1960) Possible evidence of
rapid evolution in Hawaiian moths. Evolution
14:137–138.
25 Durham WH (1992) Coevolution: Genes, Cul-
ture, and Human Diversity. Stanford: Stanford
University Press.
26 Weiner J (1995) The Beak of the Finch: A Story
of Evolution in Our Time. New York: Vintage
Books.
27 Haldane JBS (1949) Suggestions as to the
quantitative measurement of the rates of evolu-
tion. Evolution 3:51–56.
28 Grant BR, Grant PR (1993) Evolution of Dar-
win’s finches caused by a rare climatic event.
Proceedings of the Royal Society of London (B)
251:111–117.
29 Vining D (1986) Social versus reproductive
success. Behavioral and Brain Sciences 9:167–
216.
30 Williams GC (1992) Natural Selection: Do-
mains, Levels, and Challenges. Oxford: Oxford
University Press.
31 Wood JW (1994) Dynamics of Human Repro-
duction: Biology, Biometry, Demography. Haw-
thorne, NY: Aldine de Gruyter.
32 Thornhill R, Gangestad SW (1993) Human
facial beauty: Averageness, symmetry, and para-
site resistance. Human Nature 4:205–235.
33 Nesse RM, Williams GC (1994) Why We Get
Sick: The New Science of Darwinian Medicine.
New York: Times Books.
34 Gazzaniga MS (1992) Nature’s Mind: The Bio-
logical Roots of Thinking, Emotions, Sexuality,
Language, and Intelligence. New York: Basic
Books.
35 Borgerhoff Mulder M (1991) Human behav-
ioral ecology: Studies in foraging and reproduc-
tion: In Krebs JR, Davies NB (eds) Behavioral
Ecology, 3rd Edition. Oxford: Blackwell Scientific
Publications.
36 Cronk L (1991) Human behavioral ecology.
Annual Reviews of Anthropology 20:25–53.
37 Irons W (1981) Sociobiology and levels of
explanation. American Anthropologist 83:147–
149.
38 Tooby J, Cosmides L (1992) The psychological
foundation of culture. In Barkow JH, Comides L,
Tooby J (eds.) The Adapted Mind; Evolutionary
Psychology and the Generation of Culture, pp.
19–136. Oxford: Oxford University Press.
39 Daly M, Wilson M (1986) A theoretical chal-
lenge to a caricature of Darwinism. The Behav-
ioral and Brain Sciences 9:189–190.
40 Alexander RD (1989) Evolution of the human
psyche. In Mellars P, Stringer C (eds) The Human
Revolution,pp 455–513.Princeton: PrincetonUni-
versity Press.
41 McDonald K (1990) A perspective on Darwin-
ian psychology: Domain-general mechanisms,
plasticity, and individual differences. Ethol and
Sociobiol 12:449–480.
42 Irons W (1979) Cultural and biological suc-
cess. In Chagnon N, Irons W (eds.) Evolutionary
Biologyand HumanSocial Behavior: An Anthropo-
logical Perspective pp 284–302. North Scituate,
MA: Duxbury Press.
43 Irons W (1980) Is Yomut social behavior adap-
tive? In Barlow GW, Silverberg J (eds.) Sociobiol-
ogy: Beyond Nature/Nurture? pp 417–473. Boul-
der, CO: Westview Press;
44 Irons W (1997) Looking back two decades. In
Betzig L (ed) Human Nature: A Critical Reader, pp
46–49. Oxford: Oxford University Press.
45 Pe´russe D (1993) Cultural and reproductive
successin industrialsocieties: testingthe relation-
ship at the proximate and ultimate levels. Behav-
ioral and Brain Sciences 16:267–322.
46 Kitcher P (1985) Vaulting Ambition. Cam-
bridge, MA: MIT Press.
47 Rogers AR (1989) Does biology constrain cul-
ture? American Anthropologist 90:819–831.
48 Chagnon NA, Flinn MV, Melancon TF (1979)
Sex ratio variation among the Yanomamo Indi-
ans. In Chagnon NA, Irons W (eds) Evolutionary
Biologyand HumanSocial Behavior: An Anthropo-
logical Perspective, pp 290–320. North Scituate,
MA: Duxbury Press.
49 Chagnon NA (1988) Life histories, blood re-
venge,and warfare ina tribal population. Science
239:985–992.
50 Kaplan HS, Hill K (1985) Hunting ability and
reproductive success among male Ache foragers:
preliminary results. Current Anthropol 26:131–133.
51 Mealey L (1985) The relationship between
social status and biological success: a case study
of the mormon religious hierarchy. Ethol and
Sociobiol 6:249–257.
52 Borgerhoff Mulder M (1987) On cultural and
reproductive success. Amer Anthropol 89:617–634.
53 Cronk L (1991) Wealth, status, and reproduc-
tive success among the Mukogodo of Kenya.
Amer Anthropol 93:345–360.
54 Casimir MJ, Rao A (1995) Prestige, posses-
sions, and progeny. Human Nature 6:241–272.
55 Mace R (1996) Biased parental investment
andreproductive successamong Gabbrapastoral-
ists. Behav Ecol and Sociobiol 38:75–81.
56 Wadley RL (1988) Headmanship and Repro-
ductive Success among the Havasupai Indians of
Northern Arizona. M.A. Thesis, Arizona State
University.
57 PolioudakisEJ (1992) Correlates of Reproduc-
tiveSuccess over 150 Yearsin SouthernThailand.
Paper presented at the 1992 Human Behavior
and Evolution Society Meeting, Albuquerque,
NM.
58 Kaplan HS, Lancaster JB, Bock JA, Johnson
SE (1995) Does observed fertility maximize fit-
ness among New Mexican men? Human Nature
6:325–360.
59 Betzig L (1992) Roman monogamy. Ethol and
Sociobiol 13:351–384.
60 McDonald K (1990) Mechanisms of sexual
egalitarianim in Western Europe. Ethol and
Sociobiol 11:1–27.
61 Irons W (1983) Human female reproductive
strategies. In Wasser SK (ed.) Social Behavior of
Female Vertebrates, pp 169–213. New York: Aca-
demic Press.
62 Turke PW (1989) Evolution and the demand
for children. Population and Development Re-
view 15:61–90.
63 Wolf AP (1995) Sexual Attraction and Child-
hood Association: A Chinese Brief for Edward
Westermarck.Stanford: Stanford University Press.
64 Wolf AP, Huang C (1980) Marriage and Adop-
tion in China. Stanford, CA: Stanford University
Press.
65 Bischof N (1972) The biological foundations
of the incest taboo. Soc Sci Info 11:7–36.
66 van den Berghe PL, Mesher GM (1980) Royal
incest and inclusive fitness. Amer Ethnol
7:300–317.
r1998 Wiley-Liss, Inc.
204 Evolutionary Anthropology
ARTICLES