Content uploaded by Timothy Ketelaar
Author content
All content in this area was uploaded by Timothy Ketelaar on Sep 16, 2020
Content may be subject to copyright.
COMMENTARY
Clarifying the Foundations of Evolutionary Psychology:
A Reply to Lloyd and Feldman
Bruce J. Ellis
Department of Psychology
University of Canterbury, New Zealand
Timothy Ketelaar
Department of Psychology
New Mexico State University
Lloyd and Feldman’s (this issue) continuing com-
mentary on our recent target article and rejoinder (Ellis
& Ketelaar, 2000; Ketelaar & Ellis, 2000) focuses on
twooverarchingissues.First,LloydandFeldmanclaim
thatourdescriptionofthecoremetatheoreticalassump-
tions of modern evolutionary theory overemphasizes
the role of inclusive fitness (i.e., the so-called selfish
geneapproach)attheexpenseofunderemphasizingim-
portant alternative approaches (e.g., multilevel selec-
tion models, gene–culture coevolution models). Sec-
ond, Lloyd and Feldman criticize some of the methods
andassumptions that ostensiblycharacterize the evolu-
tionary psychology research program. These criticisms
concerntheconceptualizationoforganismsasinclusive
fitness maximizers, the soundness of the epistemology
of evolutionary psychology, the modularity of psycho-
logicalmechanisms,andtheuniversalityofpsychologi-
calmechanisms.Inthe firstpart of this rejoinder,we ac-
knowledge that different schools of thought exist
regarding the plausibility and importance of various
metatheoretical assumptions in human evolutionary
psychology. We argue that to date, however, only the
gene-centered adaptationist program (consistent with
inclusive fitness theory) has demonstrated scientific
progressivityby generating a coherent,integrated body
ofnewknowledgeandexplainingawayseveralapparent
anomalies. In the second part of this rejoinder, we
discuss several misunderstandings that underlie Lloyd
and Feldman’s criticisms of human evolutionary
psychology.
The Role of Inclusive Fitness Theory in
Evolutionary Psychology
Lloyd and Feldman (this issue) criticize our refer-
ence to inclusive fitness theory as providing the foun-
dation of modern evolutionary theory:
Inclusive fitness theory comprises a small subset of
models used for special purposes in evolutionary un-
derstanding. … There are many other components of
evolutionary analysis that address both animal and hu-
man behavior, and although they are conceptually
more intricate than inclusive fitness theory, they may
be more appropriate for the exploration of human psy-
chology.
Lloyd and Feldman go on to provide a technical cri-
tique of inclusive fitness theory, concluding that inclu-
sive fitness theory is of limited use and that the
foundations of evolutionary psychology are theoreti-
cally flawed.
What role does inclusive fitness theory play in evo-
lutionary psychology? As we described it (Ketelaar &
Ellis, 2000), the field of human evolutionary psychol-
ogy primarily focuses on developing and testing mid-
dle-level evolutionary theories and their derivative
hypotheses and predictions. These middle-level evolu-
tionary theories (e.g., parental investment theory, life
history theory, good genes sexual selection theory, re-
ciprocal altruism theory) are consistent with and
guided by but are not directly derived from a set of
more basic metatheoretical assumptions about
genetical evolution through natural and sexual selec-
tion (see Barkow, Cosmides, & Tooby, 1992; Buss,
1995). In this light, human evolutionary psychologists
havefoundcertain basic assumptionsabout natural and
sexual selection, such as the gene-centered
adaptationist program (consistent with inclusive fit-
ness theory), to be more useful than alternative as-
sumptions in guiding the development of middle-level
evolutionary theories; that is, in guiding the specific
theoretical models that provide a link between
metatheoretical assumptions and derivative hypothe-
ses and predictions. The armamentarium of mid-
Psychological Inquiry Copyright © 2002 by
2002, Vol. 13, No. 2, 157–164 Lawrence Erlbaum Associates, Inc.
dle-level theories employed by evolutionary
psychologists has proved useful in guiding programs
of research that have generated new knowledge of how
the mind works (Ketelaar & Ellis, 2000). At the same
time, many human evolutionary psychologists are in-
terested in investigating alternative metatheoretical as-
sumptions about natural and sexual selection, such as
multilevel selection models and gene–culture coevolu-
tion models. These alternative frameworks, however,
have not (yet) guided the development of successful
middle-level evolutionary theories and thus have not
proved useful in guiding research (as discussed later).
Lloyd and Feldman (this issue) go so far as to
claim that we did not make reference to the “real
foundations of evolutionary thought” (p. 152). At one
level of analysis, this claim is a bit like criticizing the
authors of a book on how to play billiards as having
neglected to discuss quantum mechanics as the “real
foundation” of the field of billiards. We imagine that
most books on how to play billiards adopt an ap-
proach that is more similar to Newtonian physics (the
physics of medium-size objects) than quantum me-
chanics (the physics of very small particles). Al-
though it would be incorrect to claim that quantum
mechanics is irrelevant to understanding the behavior
of billiard balls, one might argue that quantum me-
chanics is practically useless when applied to the
game of billiards. This is the case because humans do
not literally interact with billiard balls at the quantum
level. Similarly, Lloyd and Feldman may be correct
in asserting that genetical evolution is a bit more
complicated that the scenarios sketched out by Ham-
ilton (1964), Williams (1966, 1992), Dawkins (1976,
1982), and Tooby and Cosmides (1992; Cosmides &
Tooby, 1987) scenarios that lie at the heart of how
many human evolutionary psychologists view evolu-
tionary biology. Nonetheless, we argue that just as an
enjoyable (and productive) game of billiards does not
necessitate invoking insights from quantum mechan-
ics, a useful investigation of the adaptive design of
psychological mechanisms does not necessitate a di-
rect appeal to what we know about the underlying dy-
namics of genes. We are all in favor of using “the full
resources of population genetics and behavioral biol-
ogy” (Lloyd & Feldman, p.155). However, it does not
follow that because the devil is in the details of evolu-
tionary genetics that a more abstract concept such as
inclusive fitness is irrelevant to our understanding of
the behavior of medium-size biological objects such
as human psychological mechanisms.
Along these lines, Lloyd and Feldman (this issue)
endorse a different set of metatheoretical assumptions
than we endorsed in our original target article
(Ketelaar & Ellis, 2000). Specifically, Lloyd and
Feldman endorse the use of multilevel selection mod-
els and gene–culture coevolution models. Although
debate continues about the technical merits of these
different views of the evolutionary process (e.g.,
Laland, Odling-Smee, & Feldman, 2000; Wilson &
Sober, 1994), it is extraordinarily premature to sug-
gest that multilevel selection models and/or gene–cul-
ture coevolution models should replace the standard,
gene-centered adaptationist program. This is because
multilevel selection models and gene–culture co-
evolution models have (thus far) proven scientifically
barren as tools of discovery. To our knowledge, there
has not been a single new discovery about human
cognition or behavior that has been directly guided by
either multilevel selection models or gene–culture co-
evolution models. Instead, these models have at best
provided post hoc explanations of known phenomena.
This is problematic because, as Lakatos (1970, 1978)
showed, it is relatively easy to stretch existing theo-
ries to accommodate known facts. Indeed, we suspect
that there are very few empirical findings in psychol-
ogy that, after the fact, could not be claimed by multi-
ple theories as falling within their explanatory pur-
view. A good evolutionary explanation, therefore,
must not only account for known facts but also pre-
dict further, yet unobserved, novel facts (Ellis &
Ketelaar, 2000; Ketelaar & Ellis, 2000). That is, a
good explanation must stick its neck out by predict-
ing experimental results that are not known in ad-
vance.
Multilevel selection models and gene–culture co-
evolution models are complex elaborations of the
standard, gene-centered adaptationist program. What-
ever their truth value may be, these elaborations have
remained at too abstract a level to date to facilitate
scientific inquiry. Multilevel selection models and
gene–culture coevolution models are likely to be in-
corporated into the metatheory of evolutionary psy-
chology if, and when, they begin to contribute signifi-
cantly to the growth of knowledge in psychology by
anticipating new facts and resolving old puzzles. Hu-
man evolutionary psychology is a strongly empirical
field. New theoretical advances must demonstrate
their usefulness at the nuts-and-bolts empirical level
at which evolutionary psychologists conduct their
daily business. It is misleading to claim that human
evolutionary psychologists have failed to “take ad-
vantage of the full armamentarium of modern evolu-
tionary theory” (Lloyd & Feldman, p.153). The ex-
panded armamentarium that Lloyd and Feldman
recommend is simply not yet ready for combat on the
empirical battlefield.
Misunderstanding of Evolutionary
Psychology
In criticizing our depiction of the core assumptions
of human evolutionary psychology, Lloyd and
Feldman (this issue) demonstrate several misunder-
158
ELLIS & KETELAAR
standings of the field. In this section, we provide an
overview of these misunderstandings and attempt to
clarify them.
Misunderstanding #1: “Evolutionary psych-
ology claims that human behavior and its cognitive
base should be explicable in terms of enhanced in-
clusive fitness” (Lloyd & Feldman, this issue) and
“Optimization is built into inclusive fitness theory
itself; individuals are assumed to maximize their
owninclusivefitness”(Lloyd&Feldman,thisissue)
This first misunderstanding conflates the historical
process of adaptation through natural selection with
its products: adaptations themselves. The study of ad-
aptation fundamentally concerns how traits evolved
to meet the selection pressures of past, not current,
environments. Adaptations are reliably developing
characteristics of species that over evolutionarily sig-
nificant periods of time, served specific functions
(e.g., a change in color increasing the ability of an or-
ganism to avoid detection by predators) that recur-
rently promoted the survival of the genes that direct
their construction. Adaptationist analyses thus pro-
vide historical accounts of the origin and mainte-
nance of traits as a consequence of their past effects
on reproduction. The fact that a given adaptation was
produced through differential reproduction does not
imply that either (a) selection pressures are currently
favoring that adaptation or (b) that variation in the
phenotypic expression of that adaptation will be asso-
ciated with current reproductive success (Symons,
1992).
For example, the fact that correlations between so-
cioeconomic status and reproductive success are typi-
cally negative in modern Western societies (Vining,
1986) does not challenge the adaptationist hypothesis
that status striving is an evolved motivational system.
This is because the key issue in the study of adaptation
is whether differential reproductive success histori-
cally influenced the design of a given trait (e.g.,
whether high-status individuals reliably outreproduced
low-status individuals in ancestral environments) and
not whether a trait currently influences reproductive
success. The historical link between high status and re-
productive success could have been severed by any
number of novel features of modern Western environ-
ments (e.g., social welfare institutions, effective con-
traception, links between high status and years of
tertiary education).
Along these lines, a central focus of the field of
Darwinian medicine (Nesse & Williams, 1994;Wil-
liams & Nesse, 1991) is on how mismatches between
modern and ancestral environments compromise the
effectiveness of adaptations. These mismatches often
result in dysfunctional behavior that is inexplicable in
terms of enhanced inclusive fitness. For example, the
dopamine-mediated reward mechanisms found in the
mesolimbic system in the brain appear to have
evolved to provide a pleasurable reward in the pres-
ence of adaptively relevant stimuli such as food or
sex. In contemporary environments, however, these
same mechanisms are subverted by the use of psy-
choactive drugs such as cocaine and amphetamines,
which deliver huge dollops of pleasurable reward in
the absence of the adaptively relevant stimuli, often
to the user’ s detriment (Nesse & Berridge, 1997). In
total, evolutionary psychologists (e.g., Buss, 1995;
Symons, 1992; Tooby & Cosmides, 1992) strongly
reject the notion that “human behavior and its cogni-
tive base should be explicable in terms of [currently]
enhanced inclusive fitness” (Foley, 1995–1996, as
cited in Lloyd & Feldman, this issue) and that “indi-
viduals are assumed to maximize their own inclusive
fitness” (Lloyd & Feldman, p.155).
Misunderstanding #2: The methods used to
generate and test evolutionary psychological
models and hypotheses are not scientifically
defensible (Lloyd & Feldman, this issue)
Lloyd and Feldman (this issue) assert that the
methods employed by evolutionary psychologists to
generate and test evolutionary explanations are “sim-
ply not good enough to be regarded as part of the sci-
ence of evolutionary biology” (p.155). Given that the
scientific defensibility of evolutionary psychological
explanations was the central focus of our original tar-
get article, we are quite puzzled that Lloyd and
Feldman revisit the “bad science” criticism without
engaging any of our original arguments on the topic.
We reiterate that none of the original 16 commenta-
tors on our target article challenged our central thesis:
that the methods and strategies employed by evolu-
tionary psychologists to generate and test hypotheses
are scientifically defensible. In this regard, Lloyd and
Feldman offer nothing new by raising this criticism
afresh.
A common theme in criticisms of evolutionary
psychological explanations is that they are unverifi-
able because they make inferences about historical
processes that cannot be directly observed (e.g.,
Lewontin, 1998; Lloyd, 1999; Richardson, 1996,
2000). Brandon (1990; see also Richardson, 1996)
suggested that one cannot claim to know how a trait
actually evolved unless one acquires the following
five sources of information: (a) evidence that selec-
tion has occurred (i.e., that the trait of interest is not
the product of other evolutionary forces), (b) evi-
dence that there is some ecological factor in the envi-
ronment that offers an explanation for the strength of
selection, (c) evidence that the trait has a genetic ba-
sis, (d) evidence concerning the structure of the popu-
159
COMMENTARY
lation both from a genetic and selective viewpoint,
and (e) phylogenetic information concerning primi-
tive and derived characteristics. Lacking such evi-
dence, one is restricted to offering accounts of how a
trait might possibly have evolved. Brandon acknowl-
edged that these five criteria are rarely met in
adaptationist explanations for any characteristic in
any species. Lloyd and Feldman (this issue) cite the
recent evolution of the gene for human sickle cell dis-
ease in response to the malarial parasite as an exam-
ple that does begin to meet these kind of verification
criteria. No evolutionary explanation of any trait of
interest to psychologists, however, is likely to meet
all of these criteria. Accordingly, evolutionary psy-
chological explanations are seen as being based
largely on claims of plausibility and, as a result, ap-
pear unverifiable on evidential grounds (e.g., Davies,
1996; Lewontin, 1998; Lloyd, 1999; Richardson,
1996). The charge is that evolutionary psychological
explanations are, therefore, of little scientific value.
As we argued previously (see Ellis & Ketelaar,
2000), there are at least two important qualifications to
this criticism that must be appreciated when evaluating
evolutionary psychological explanations. First, unveri-
fiability is conceptually distinct from unfalsifiability.
As articulated by Popper (1959), although particular
statements (hypotheses) cannot be unequivocally veri-
fied, they can be refuted or falsified. For this reason,
Popper placed corroboration (establishing that an ex-
planation is consistent with the data) rather than verifi-
cation in opposition to falsification. Popper argued that
although it is possible to gather corroborating data to
support a particular explanation, no amount of em-
pirical evidence would allow, for example, Sherlock
Holmes to unequivocally verify (i.e., prove) a particu-
lar explanation for the discovery of a cadaver (e.g.,
“the butler committed the murder with a pistol in the
pantry”). This is because it is almost always possible
to identify alternative inductive inferences that are
consistent with the existing corpus of data. This is es-
pecially true of explanations that specify unobservable
entities and processes (Churchland, 1989). Thus, in
modern philosophy of science, verification is generally
considered too high a standard to be of practical use in
evaluating scientific explanations (e.g., Feigl, 1956;
Haig & Durrant, 2000; Hooker, 1987; Meehl, 1990;
Salmon, 1967, 1990). Nonetheless, a single piece of
evidence (e.g., the fact that the maid had been dead for
3 years) could be used to unequivocally reject an alter-
native explanation (e.g., “the murder was committed
by the maid”). In science, then, just as in the judicial
system, certain accounts are accepted over alternative
accounts, not because they have been absolutely veri-
fied (shown to be true), but rather because they alone
are the best available explanation that emerges from
the set of corroborated alternative accounts.
Rather than emphasizing verification, a more com-
mon approach to science emphasizes the process of es-
tablishing acceptable claims to knowledge through
methods analogous to construct validation (e.g.,
Cronbach & Meehl, 1955; Hooker, 1987; Meehl,
1990). To establish construct validity, it is necessary
for a middle-level evolutionary theory to specify the
meaning of a construct (i.e., to specify the functional
design of a proposed adaptation), distinguish it from
other constructs, and generate clear, testable hypothe-
ses and predictions about how measures of the pro-
posed construct should and should not relate to other
variables (see the American Psychological Associa-
tion’s, 1985, discussion of construct validation). If a
prediction is supported, this provides indirect support
forthe hypotheses, whichin turn, mayprovide grounds
for favoring one theoretical model over another. If a
prediction is disconfirmed, it is back to the drawing
board, either to attempt a better translation of the hy-
pothesis (into a specific prediction) or to actually mod-
ify or reject the hypothesis altogether (Ketelaar &
Ellis, 2000). This process may, in turn, lead to revision
of the middle-level evolutionary theory from which the
hypothesis was derived or to replacement of the mid-
dle-level theory by a plausible alternative. In this man-
ner, well-validated evolutionary explanations can
count as acceptable knowledge claims; that is, these
explanations can be empirically corroborated (or falsi-
fied) relative to other explanations, even though they
cannot be directly verified (see Salmon, 1967, 1990,
1992, for a more detailed discussion of verification and
inductive reasoning in science).
Second, even if an evolutionary psychological
model has experienced predictive failures or falsifica-
tions and even if the model fails to meet Brandon’s
(1990) five criteria for verification, this does not itself
constitute adequate grounds for rejecting the model.
As many philosophers of science have noted (e.g.,
Lakatos, 1978; Thagard, 1992), theory evaluation is
not simply a matter of gauging the degree of fit be-
tween a single theory and the world. Rather, theories
are evaluated in relation to alternative theories based
on the cumulative weight of the evidence. Acceptance
or rejection of a given evolutionary psychological ex-
planation depends on comparison with other explana-
tions of the trait in question. It is not enough to claim
that a given evolutionary psychological explanation is
“highly unscientific or unverified” (Lloyd & Feldman,
p.155). Rather, substantive critiques must demonstrate
that alternative explanations (whether adaptationist or
nonadaptationist in origin) provide a better account of
the trait in question than does the proposed evolution-
ary psychological explanation. Lloyd and Feldman
(this issue) offer little by way of empirically corrobo-
rated alternatives to the basic assumptions that they
call into question.
160
ELLIS & KETELAAR
Misunderstanding #3: “Evolutionary psycholo-
gists assume that human beings have a module (or
set of genes) for each task that is imposed by the en-
vironment” (Lloyd & Feldman, this issue)
Although there is not consensus among evolution-
ary psychologists concerning the level of domain
specificity of psychological mechanisms, the notion
that “human beings have a module (or set of genes)
for each task that is imposed by the environment”
(Lloyd & Feldman, p. 153) is at best a minority posi-
tion in the field. It is certainly not a position that we
endorsed in our original target article (Ketelaar &
Ellis, 2000) or rejoinder (Ellis & Ketelaar, 2000). The
assumption that “the human brain/mind is comprised
of a large number of specialized cognitive adaptations
that were shaped by natural selection over vast peri-
ods of time to solve the recurrent information-pro-
cessing problems faced by our ancestors” (Ellis &
Ketelaar, 2000, p. 61) is widely endorsed by evolu-
tionary psychologists. This assumption, however,
does not imply a one-to-one correspondence between
number of information-processing problems and
number of specialized cognitive adaptations. As dis-
cussed in our original rejoinder, there is ongoing de-
bate among evolutionary and cognitive psychologists
concerning the numerosity and specificity of psycho-
logical mechanisms.
Samuels (2000) usefully distinguished between
strong and weak versions of the massive modularity
hypothesis. According to the strong version, the hu-
man brain–mind is composed entirely of Darwinian
modules: innate, special-purpose computational
mechanisms that were designed by natural and sexual
selection. According to the weak version, the human
brain–mind is composed largely but not entirely of
Darwinian modules. The weak version of the massive
modularity hypothesis is consistent with the assump-
tion that different psychological mechanisms differ in
their levels of specificity and that there are higher
level executive mechanisms that operate on informa-
tion drawn from a range of more domain-specific
lower level mechanisms. Evolutionary psychologists
have only committed themselves to the weak version
of the massive modularity hypothesis (Samuels,
2000). The weak version, for example, was clearly
endorsed by Tooby and Cosmides (1998):
The deepest issues in cognitive science involve the
heterarchical and cross-cutting scope of different
mechanisms: what domains are native to the human
mind, what is the mixture of domain-specific and do-
main-general devices that compute over these do-
mains, what are the designs of these devices, and how
do they pass inputs and outputs back and forth. These
same issues apply to human anatomy and physiology:
even tissues of very different organs employ identical
processes at some levels (RNA transcription, mito-
chondrial energy production) and highly specialized
processes at others (hemoglobin binding for oxygen
transport). … Sorting this out at an information pro-
cessing level will be a fascinating task, and the answer
will not be reducible to “Everything is domain-spe-
cific” or “Everything is domain-general.” (p. 200)
We concur with Lloyd and Feldman (this issue; cf.
Shapiro & Epstein, 1998) that selection is unlikely to
favor distinct Darwinian modules for the solution of
every adaptive problem. This would be inefficient,
given that some modules can conceivably contribute to
the solution of two or more adaptive problems. Indeed,
there is clear evidence of multiplicity of function in
certain noncognitive traits (e.g., bird feathers function
both to regulate temperature and facilitate flight).
There should also be multiplicity of function in certain
cognitive traits (e.g., input of frequentist mechanisms
to many types of judgments; Brase, Cosmides, &
Tooby, 1998).
We also concur with Lloyd and Feldman (this
issue) that there are real evolutionary constraints on
the design features of adaptations. Rather than de-
scribing Darwinian modules as optimally designed
(from the standpoint of a rational engineer trying to
solve a problem), evolutionary psychologists more
typically use terms such as gerrymandered devices,
Rube–Goldberg machines, or weird contraptions to
describe adaptations (see Dawkins, 1982; Dennett,
1995). This is because evolutionary psychologists ac-
knowledge many possible constraints on optimal de-
sign, such as the slow speed of evolution, the lack of
available genetic variation, the need to build on pre-
existing structures, and the necessity of coordination
with other mechanisms (e.g., Buss et al., 1998;
Dawkins, 1982).
Misunderstanding #4: High within-population
variability on most measured human behavioral
traits undermines the assumption of universality of
psychological mechanisms, that is, the assumption
that there is a species-wide evolved human nature
(Lloyd & Feldman, this issue)
High within-population variability on most human
behavioral traits does not, in and of itself, challenge the
assumption of an innate, universal human nature. Uni-
versally shared, functionally specialized cognitive ad-
aptationscan generate individualdifferencesas a result
of different life experiences. Rather, it is the existence
of heritable variation in human behavioral traits that
calls into question the idea of a universal human nature
(Tooby & Cosmides, 1990). Attempts to explain heri-
161
COMMENTARY
table variation is an active area of theory and research
in evolutionary psychology, and a number of different
viewpoints have been expressed on the topic.
Tooby and Cosmides(1990), for example, acknowl-
edged that vast reservoirs of genetic variability under-
lie human psychological functioning but argued that
this variability is “generally limited to quantitative
variation in the components of complex, highly articu-
lated, species-typical psychological mechanisms” (p.
24). According to Tooby and Cosmides, pathogen
pressures select for multiple-alternative alleles at a
large proportion of loci, introducing as much quantita-
tive variation and noise into adaptive systems as can be
tolerated without compromising their functional integ-
rity. From this perspective, the genetic component of
phenotypic variation mostly constitutes genetic noise
(nonadaptive random fluctuations around species-typi-
cal design).
Other evolutionary psychologists, however, have
challenged the genetic noise argument. Miller (2000a,
2000b), for example, posited that heritable variation
in sexually selected fitness indicators is adaptively
patterned. A core premise of good genes sexual selec-
tion theory is that certain traits have evolved because
they are reliable indicators of genetic quality; that is,
these traits reliably signal viability and good condi-
tion that can be passed on to offspring through ge-
netic inheritance. For a given trait to be a reliable
indicator of genetic quality, it must be costly to pro-
duce. According to the handicap principle (Zahavi &
Zahavi, 1997), sexually selected fitness indicators,
such as the peacock’s tail, can only be maintained by
individuals who are the fittest in the population, as in-
dicated by their ability to maintain steady growth
rates, resist parasites, compete successfully in intra-
sexual contests, and so forth. Consequently, individu-
als who more fully display sexually selected fitness
indicators tend to be healthier, in better condition,
and more preferred as mates than conspecifics who
display these traits less fully (e.g., Gangestad &
Simpson, 2000; Moller, Christie, & Lux, 1999).
Kotiaho, Simmons, and Tomkins (2001) demon-
strated that there are high levels of additive genetic
variance underlying individual differences in condi-
tion (operationalized as the ability to withstand nutri-
tional stress) and that strong positive genetic correla-
tions exist between condition and courtship rates.
Pomiankowski and Moller (1995) suggested that
adaptive genetic variation in sexually selected fitness
indicators is maintained by long-term directional
selection.
The genetic noise argument was also challenged by
Bussand Greiling (1999),who argued that heritablein-
dividualdifferencesinhumanpersonalitytraitsarereli-
ably linked to reproductively relevant phenomena such
assexualbehaviorandstatus attainment. Over the last2
decades, theory and research in evolutionary biology
hasbegun to acknowledgethatin most species,a single
“best” reproductive strategy is unlikely to evolve
(Gangestad& Simpson, 2000; Gross,1996).This is be-
causethe best reproductivestrategyvariesasa function
of the physical, economic, and social parameters of
one’senvironment(Crawford&Anderson, 1989).Heri-
table variation in personality traits reliably guides indi-
viduals toward different reproductive strategies (e.g.,
Bailey & Martin, 2000; Eysenck, 1976). Some of this
variationmay,thus,beproducedandmaintainedbynat-
ural and sexual selection.
Along these lines, evolutionary psychologists have
begun to develop and test frequency-dependent models
of individual differences in personality. Heritable vari-
ation in personality can be produced and maintained at
equilibrium in populations through frequency-depend-
entselection (where thefitnessof a heritablevariantin-
creases as it becomes rarer in the population;
frequency-dependent selection, thus, favors parents
who produce more offspring of the rare variant).
Gangestad and Simpson (1990) have proposed that
heritable variation in women’s sociosexual orientation
(i.e., women’s orientation toward long-term versus
short-term mating) is the result of frequency-depend-
ent selection (see Bailey & Martin, 2000, for a recent
empirical test of this model). Likewise, Mealey (1995)
suggested that heritable variation in primary socio-
pathy is the result of frequency-dependent selection.
It is important to note that even if good genes sexual
selection produces adaptive variation in fitness indica-
tors and even if some heritable variation in personality
traits is maintained by frequency-dependent selection,
this does not undermine the assumption that there is a
species-wide evolved human nature. This is because
universallysharedpsychological adaptations cancoex-
ist with other adaptations that are expressed in alterna-
tive forms in different individuals. One does not
exclude the other.
Conclusions: Two Schools of Thought
in the Evolutionary Sciences
We conclude by placing the debate between our-
selves and Lloyd and Feldman (this issue) in the con-
text of a larger, ongoing debate in the evolutionary
sciences. Sociologist Ullrica Segerstrales (2000) re-
cently examined the last several decades of the
so-called sociobiology debate and concluded that this
debate often boils down to a conflict between two main
schools of thought: the critical experimentalist school
and the naturalist school. Many of Lloyd and
Feldman’s concerns appear to reflect the perspective of
the critical experimentalist school of thought, whereas
muchof the adaptationistprogram in humanevolution-
ary psychology, which we have described, reflects the
naturalist school of thought.
162
COMMENTARY
According to Segerstrales (2000), the critical
experimentalist school is comprised largely of geneti-
cists, biologists, and palaeontologists (e.g., Lewontin,
Rose, Gould) whose work is largely confined to the
laboratory (thus, the moniker experimentalists). These
scientists work with actual physical constructs such as
alleles, cells, and fossils. By virtue of their extensive
focus on laboratory techniques, individuals trained in
this school of thought tend to emphasize the complica-
tions and caveats involved in defining, measuring, and
conceptualizing abstract constructs such as genes and
adaptations. From the perspective of the critical
experimentalist school of thought, human evolutionary
psychologists often take great license in applying our
laboratory knowledge of genes, developmental embry-
ology, and the fossil record to the “real” world of
minds, muscles, and behavior. By contrast, the natural-
istschool of thoughtis comprised largelyofzoologists,
behavioral ecologists, evolutionary psychologists, and
biologists (e.g., Dawkins, Alexander, Williams, Cos-
mides,Tooby,Buss)who are lessinterested in studying
genes, cells, and fossils in the laboratory and much
more interested in studying the behavior of organisms
in their natural settings. Although proponents of the
naturalist school are generally aware of the caveats in-
volved in defining and measuring basic concepts such
as genes, they are not as reluctant (as the critical
experimentalists) to posit “hypothetical” constructs
such as “evolved psychological mechanisms.” For in-
dividuals trained in the naturalist school of thought, the
admittedly abstract and onerous concept of adaptation
is a much more natural level of description (than
cistrons, alleles, and proteins) for understanding the
properties of entire organisms observed in their natural
environments.
It is not surprising that critical experimentalists,
who have made important contributions to our under-
standing of the microworld of genetics, would call on
naturalists to be more cognizant of assumptions at the
microlevel when investigating macrolevel entities and
processes. After all, a complete understanding of our
evolved human nature necessarily entails knowledge
spanning from the microlevel of genes and proteins to
the macrolevel of human behavior and cognition. True
progress in the human evolutionary sciences will occur
when these very different levels of analysis are suc-
cessfully integrated.
Notes
Bruce J. Ellis, Department of Psychology, Univer-
sity of Canterbury, Private Bag 4800, Christchurch,
New Zealand. E-mail: b.ellis@psyc.canterbury.ac.nz.
Timothy Ketelaar, Department of Psychology, New
Mexico State University, Las Cruces, NM 88003.
E-mail: ketelaar@comstds.ucla.edu
References
American Psychological Association. (1985). Standards for educa-
tional and psychological tests. Washington, DC: Author.
Bailey, J. M., Kirk, K. M., Zhu, G., Dunne, M. P., & Martin, N. G.
(2000). Do individual differences in sociosexuality represent
genetic or environmentally contingent strategies? Evidence
fromtheAustraliantwinregistry.JournalofPersonalityandSo-
cial Psychology, 78(3), 537–545.
Barkow, J. H., Cosmides, L., & Tooby, J. (Eds.). (1992). The adapted
mind: Evolutionary psychology and the generation of culture.
New York: Oxford University Press.
Brandon, R. N. (1990). Adaptation and environment. Princeton, NJ:
Princeton University Press.
Brase, G. L., Cosmides, L., & Tooby, J. (1998). Individuation,
counting, and statistical inference: The role of frequency
and whole-object representations in judgment under uncer-
tainty. Journal of Experimental Psychology: General, 127,
3–21.
Buss, D. M. (1995). Evolutionary psychology: A new paradigm for
psychological science. Psychological inquiry, 6, 1–30.
Buss, D. M., & Greiling, H. (1999). Adaptive individual differences.
Journal of Personality, 67, 209–243.
Buss, D. M., Haselton, M. G., Shackelford, T. K., Bleske, A. L., &
Wakefield, J. C. (1998). Adaptations, exaptations, and span-
drels. American Psychologist, 53, 533–548.
Churchland, P. M. (1989). A neurocomputational perspective: The
nature of mind and the structure of science. Cambridge, MA:
MIT Press.
Cosmides,L.&Tooby,J.(1987).Fromevolutiontobehavior:Evolu-
tionary psychology as the missing link. In J. Dupré (Ed.), The
latest on the best: Essays on evolution and optimality (pp.
277–306). Cambridge, MA: MIT Press.
Crawford, C. B., & Anderson, J. L. (1989). Sociobiology: An envi-
ronmentalist discipline? American Psychologist, 44,
1449–1459.
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psycho-
logical tests. Psychological Bulletin, 52, 281–302.
Davies, P. S. (1996). Evolutionary theory in cognitive psychology.
Minds and Machines, 6, 445–462.
Dawkins, R. (1976). The selfish gene. New York: Oxford University
Press.
Dawkins, R. (1982). The extended phenotype. San Francisco: Free-
man.
Dawkins, R. (1986). The blind watchmaker. New York: Norton.
Dennett, D. C. (1995). Darwin’s dangerous idea: Evolution and the
meanings of life. New York: Simon & Schuster.
Ellis, B. J., & Ketelaar, T. (2000). On the natural selection of alterna-
tive models: Evaluations of explanations in evolutionary psy-
chology. Psychological Inquiry, 11, 56–68.
Eysenck, H. J. (1976). Sex and personality. Austin: University of
Texas Press.
Feigl,H.(1956).Somemajor issues and developmentsinthephiloso-
phy of science of logical empiricism. In H. Feigl & M. Scriven
(Eds.), The foundations of science and the concepts of psychol-
ogy and psychoanalysis (pp. 3–37). Minneapolis: University of
Minnesota Press.
Gangestad, S. W., & Simpson, J. A. (2000). The evolution of human
mating: Trade-offs and strategic pluralism. Behavioral and
Brain Sciences, 23, 573–644.
Gangestad, S. W., & Simpson, J. S. (1990). Toward an evolutionary
history of female sociosexual variation. Journal of Personality,
58, 69–96.
Gross, M. R. (1996). Alternative reproductive strategies and tactics:
Diversity within sexes. Trends in Ecology and Evolution, 11,
92–98.
Haig, B. D., & Durrant, R. (2000). Theory evaluation in evolutionary
psychology. Psychological Inquiry, 11, 34–38.
163
COMMENTARY
Hamilton, W. D. (1964). The genetical evolution of social behavior.
Journal of Theoretical Biology, 7, 1–52.
Hooker, C. A. (1987). A realistic theory of science. Albany: State
University of New York Press.
Ketelaar, T., & Ellis, B. J. (2000). Are evolutionary explanations
unfalsifiable? Evolutionary psychology and the Lakatosian phi-
losophy of science. Psychological Inquiry, 11, 1–21.
Kotiaho, J. S., Simmons, L. W., & Tomkins, J. L. (2001). Towards a
resolution of the lek paradox. Nature, 410, 684–686.
Lakatos,I.(1970).Falsificationismandthemethodologyof scientific
researchprogrammes.InI.Lakatos&A.Musgrave (Eds.), Crit-
icism and the growth of knowledge (pp. 91–196). Cambridge,
England: Cambridge University Press.
Lakatos, I. (1978). The methodology of scientific research
programmes: Philosophical papers (Vol. 1, J. Worral & G. Cur-
rie, Eds.). Cambridge, England: Cambridge University Press.
Laland, K. N., Odling-Smee, J., & Feldman, M. W. (2000). Niche
construction, biological evolution, and cultural change. Behav-
ioral and Brain Sciences, 23, 131–175.
Lewontin, R. C. (1998). The evolution of cognition. In D.
Scarborough & S. Sternberg (Eds.), An invitation to cognitive
science: Methods, models, and conceptual issues (pp.
107–132). Cambridge, MA: MIT Press.
Lloyd,E.A.(1999).Evolutionarypsychology:Theburdens of proof.
Biology and Philosophy, 14, 211–233.
Mealey, L. (1995). The sociobiology of sociopathy: An integrated
evolutionary model. Behavioral and Brain Sciences, 18,
523–599.
Meehl, P. E. (1990). Appraising and amending theories: The strategy
of Lakatosian defense and two principles that warrant it. Psy-
chological Inquiry, 1, 108–141.
Miller, G. (2000a). The mating mind: How sexual choice shaped the
evolution of human nature. London: Heinemann.
Miller, G. (2000b). Mental traits as fitness indicators: Expanding evo-
lutionary psychology’s adaptationism. In D. LeCroy & P. Moller
(Eds.) Proceedings of the New York Academy of Sciences: Evolu-
tionary perspectives on human reproductive behavior (Vol. 907,
pp. 2–74). New York: New York Academy of Sciences.
Moller, A. P., Christie, P., & Lux, E. (1999). Parasitism, host immune
function, and sexual selection. Quarterly Review of Biology, 74,
3–20.
Nesse,R.M.,& Berridge, K.C.(1997).Psychoactivedrugusein evo-
lutionary perspective. Science, 278, 63–66.
Nesse, R. M., & Williams, G. C. (1994). Why we get sick: The new
science of Darwinian medicine. New York: Times Books.
Pomiankowski,A.,& Moller,A.P.(1995).Aresolution of thelekpar-
adox.ProceedingsoftheRoyalSociety ofLondonB,260,21–29.
Popper, K. R. (1959). The logic of scientific discovery. New York:
Hutchison Education.
Richardson, R. C. (1996). The prospects for a human evolutionary
psychology:Humanlanguageandhumanreasoning.Mindsand
Machines, 6, 541–557.
Richardson,R.C.(2000).Epicyclesandexplanationsinevolutionary
psychology. Psychological Inquiry, 11, 46–49.
Salmon, W. C. (1967). The foundations of scientific inference. Pitts-
burgh, PA: University of Pittsburgh Press.
Salmon, W. C. (1990). Rationality and objectivity in science, or Tom
Kuhn meets Tom Bayes. In C.W. Savage (Ed.), Scientific theo-
ries,Minnesotastudiesinphilosophyofscience,(pp.175–204).
Minneapolis: University of Minnesota Press.
Salmon, W. C. (1992). Epistemology of natural science. In J. Dancy
& E. Sosa (Eds.), A companion to epistemology (pp. 292–296).
Oxford, England: Blackwell.
Samuels, R. (2000). Massively modular minds: Evolutionary psy-
chologyandcognitivearchitecture. In P.Carruthers & A. Cham-
berlain (Eds.), Evolution and the human mind: Modularity, lan-
guage and meta-cognition (pp. 13–46). Cambridge, England:
Cambridge University Press.
Segerstrales,U.(2000).Defendersofthetruth:Thebattleforscience
in the socio-biology debate and beyond. Oxford, England: Ox-
ford University Press.
Shapiro, L., & Epstein, W. (1998). Evolutionary theory meets cogni-
tive psychology: A more selective perspective. Mind and Lan-
guage, 13, 171–194.
Symons, D. (1992). On the use and misuse of Darwinism in the study
of human behavior. In J. H. Barkow, L. Cosmides, & J. Tooby
(Eds.), The adapted mind: Evolutionary psychology and the gen-
eration of culture (pp. 137–159). New York: Oxford University
Press.
Thagard,P. (1992). Conceptual revolutions.Princeton,NJ:Princeton
University Press.
Tooby, J., & Cosmides, L. (1990). On the universality of human na-
ture and the uniqueness of the individual: The role of genetics
and adaptation. Journal of Personality, 58, 17–67.
Tooby, J., & Cosmides, L. (1992). The psychological foundations of
culture. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The
adapted mind: Evolutionary psychology and the generation of
culture (pp. 19–136). New York: Oxford University Press.
Tooby, J., & Cosmides, L. (1998). Evolutionizing the cognitive sciences:
AreplytoShapiro and Epstein.MindandLanguage,13, 195–204.
Vining,D.R.(1986).Socialversusreproductive success: The central
theoretical problem of human sociobiology. Behavioral and
Brain Sciences, 9, 167–187.
Williams, G. C. (1966). Adaptation and natural selection. Princeton,
NJ: Princeton University Press.
Williams,G.C.(1992). Natural selection:Domains,levels,andchal-
lenges. New York: Oxford University Press.
Williams, G. C., & Nesse, R. M. (1991). The dawn of Darwinian
medicine. Quarterly Review of Biology, 66, 1–22.
Wilson, D. S., & Sober, E. (1994). Reintroducing group selection to
the human behavioral sciences. Behavioral and Brain Sciences,
177, 585–654.
Zahavi, A., & Zahavi, A. (1997). The handicap principle: A missing
piece of Darwin’s puzzle. New York: Oxford University Press.
164
ELLIS & KETELAAR
