| |

Exchange

Genes and Ideologies

Evan Charney

There is a trend among behavioral scientists to view ever more complex attitudes or systems of belief as in some sense
genetically determined (or “heritable”). Consistent with this trend is the recent article of Alford, Funk, and Hibbing titled “Are
Political Orientations Genetically Transmitted?” in which the authors claim to have demonstrated that when it comes to the
transmission of political ideologies, genes count for more than environment. Their article has received an enormous amount of
attention among political scientists and in the popular press. I critically evaluate the research technique on the basis of which
the authors’ support their claims and argue that it suffers from significant methodological flaws. Such flaws notwithstanding, I
demonstrate that the authors’ data do not clearly support their conclusions. I then question the cogency, from an historical and
theoretical perspective, of proposing the existence of “liberal” and “conservative” “phenotypes” and “genotypes.” My argument
has implications beyond the findings of Alford, Funk, and Hibbing, and applies to all studies that claim to have demonstrated
the heritability of complex and politically relevant attitudes.

Researchers in human behavioral genetics have taken an overly simplistic view of human social behaviors and
aptitudes. The failure to consider more complex views of the interactions both between genes and between genes
and environment may explain the absence of any bona fide findings of genes associated with the behaviors they
study. Nevertheless, the studies have often been presented by scientists as if conclusive, thus attracting consider-
able media and public attention. Journalists have moved from conclusions based upon the simplistic assump-
tions underlying the studies to present to the public simplistic views of human social problems and human social 䡬
arrangements. This translation is sometimes encouraged by scientists themselves, providing journalists with
provocative statements for their newspapers and magazines.1

here is a growing trend among behavioral scientists regained the ascendancy it once held when William Gal-

T (particularly psychologists) and now, political scien-

tists, to view more and more of human behavior as
in large measure attributable to our genes. In the old
debate of “nature v. nurture,” nature now seems to have
Evan Charney is Assistant Professor of Public Policy and
Political Science at Duke University (echar@duke.edu). An
earlier version of this paper was presented as part of a panel
on Genetic and Evolutionary Bases of Political Behavior at
the 2006 APSA annual meeting. He would like to thank
Jonathan Beckwith and Cory A. Morris for agreeing to
provide a written response to Alford, Funk, and Hibbing
and Hannagan and Hetemi; Tracy Rupp for her many
invaluable comments and suggestions; Ira Morgan for
helpful exchanges, and the members of the Duke Institute
for Genome Sciences and Polich and the IGSP Center for
Genome Ethics, Law, and Policy, and in particular Robert
Cook-Deegan and William English. He would especially
like to thank Jim Johnson for his clear commitment to
genuine debate within the profession as evidenced by his
willingness to publish an article that runs counter to the
latest vogue in political science.
ton, author of Hereditary Genius, wrote in 1875 after one
of the first “twin studies,” that as regards behavioral,
psychological, and cognitive characteristics, “There is
no escape from the conclusion that nature prevails enor-
mously over nurture.” 2 A popular book on genetics asserts
triumphantly (on behalf of nature): “Genes not only deter-
mine how we look, but how we act, feel, and experience
life. In case after case discovered by researchers, nature
won out over nurture.” 3 Almost every month a new study
appears purporting to show that ever more complex aspects
of human intellect, belief, and behavior—everything from
intelligence (i.e., IQ) to the level of such complex atti-
tudes or beliefs as altruism, criminality, religiosity, and
conservatism—can be explained largely, and in some
instances exclusively, by an individual’s genetic makeup.4
Consistent with this trend to view complex aspects of
human belief as “heritable” is the argument recently pre-
sented by Alford, Funk, and Hibbing in their article, “Are
Political Orientations Genetically Transmitted?” 5 In this
article, the authors claim to have demonstrated that the
answer to this question is unambiguously yes, and that
“political attitudes are influenced much more by genetics
than by parental socialization.” 6 Specifically, they claim to

doi:10.1017/S1537592708080626 June 2008 | Vol. 6/No. 2 299

䡬
| |
Exchange | Genes and Ideologies
have discovered that when it comes to political conserva-
tism, “genetics accounts for approximately half of the vari-
ance in ideology, while shared environment including
parental influence accounts for only 11%.” 7 Understand-
ably, this astonishing claim has garnered a great deal of
attention among both political scientists and in the pop-
ular press. As the editor of the American Political Science
Review has recently commented, the 2005 article of Alford,
Funk, and Hibbing, “has set a new standard for political
science in terms of the media attention and public discus-
sion that its publication has provoked,” and he goes on to
wonder “whether it will emerge among the most impor-
tant articles the APSR has ever published.” 8
The claim of Alford, Funk, and Hibbing is indeed aston-
ishing, because if true, it would require nothing less than
a revision of our understanding of all of human history,
much—if not most—of political science, sociology, anthro-
pology, and psychology, as well as, perhaps, our under-
standing of what it means to be human. True to Beckwith’s
quote (cited above), the study of Alford, Funk, and Hib-
bing (hereafter AFH) has been presented by the authors as
conclusively demonstrating that political orientations are
genetically transmitted, and most of the press coverage
has accepted the authors’ claims at face value.9 As yet,
there has been no critical response to the authors’ article
in any scholarly journals (or in the press), and given the
䡬 enormous attention their claims have received, and the
extraordinary nature of the claims themselves, such a
response is sorely needed. It is my objective in this paper
to present such a response. In sum, my argument is that
the authors have in no way demonstrated their hypothesis
concerning the heritability of political ideologies, and that
such a hypothesis is, in and of itself, extremely implausi-
ble (if not incoherent). Although this paper focuses on the
article of AFH, as should be clear, most of the criticisms
presented here have a much broader applicability, and could
apply equally to many recent claims regarding the herita-
bility of complex human beliefs and attitudes.
My argument appears in five sections. First, I present
some brief comments concerning the concept of herita-
bility, a concept which is easily (and commonly) misun-
derstood. Next, I critique the research technique on the
basis of which AFH make their claims. I then consider the
data presented by AFH and the extent to which it sup-
ports the authors’ claims. Following this, I examine a num-
ber of issues relating to liberalism and conservatism as
political ideologies, both historical and conceptual, and
examine the conceptual plausibility of assumptions con-
cerning the heritability of specific political beliefs.
This article is eclectic, intentionally so., I present argu-
ments referring to genetics, biology, neurology, research
methodology, the nature and history of liberalism and
conservatism as political ideologies, political psychology,
and the nature of morality and reason. It is my hope here
not simply to open up a dialogue on the subject of the
300 Perspectives on Politics
䡬
relationship, if any, between genes and political ideology,
but to open up a dialogue that can effectively cross sub-
discipline boundaries within political science itself, and
perhaps even spark a much-needed public dialogue on
this subject.
I would like to emphasize that many of the claims con-
cerning the heritability of complex human beliefs and atti-
tudes have obvious and profound policy implications. For
example, claims concerning the heritability of criminal
behavior and the existence of a so-called “criminal gene”
have profound implications for the criminal justice sys-
tem.10 According to Lindsay Elkins, “the discoveries of
the Human Genome Project are already shaking the foun-
dations of our legal system, particularly in the area of
criminal law.” 11 Claims concerning the heritability of IQ
have profound implications for how we view the aims of
public education, as well as questions of social and eco-
nomic justice more generally. The controversy sparked by
the claims of Herrnstein and Murray in The Bell Curve
concerning the supposed relationship between IQ—which
the authors assume to be largely heritable—and race, is
well known to all social scientists.12 While no contempo-
rary author, to my knowledge, has discussed in any detail
the policy implications of an assumption that political
ideology is heritable, they are profound indeed (consider
the political ideology of, e.g., a member of al-Qaeda).
The Meaning of Heritability
Heritability is an estimate of the amount of variation in
a given trait (phenotype) in a given population at a given
time that is due to genes (genotype). A heritability estimate
of .50, for example, does not mean that for any given indi-
vidual 50 percent of that trait is due to or caused by one’s
genes. Rather, it means that 50 percent of the variation in a
given trait across an entire population at a given time is the
result of differences in persons’ genotypes. Heritability is
thus a population-specific estimate; estimates of heritabil-
ity within a given population do not indicate the causes of
phenotypic variation between populations.
It is a common misunderstanding that estimates of her-
itability are fixed and invariable, usually based on the
assumption that they reveal something fixed and inevita-
ble (or deterministic) about human biology (as expressed
through human genes). To the contrary: The heritability of
a trait can change with changes in the environment. In a
classic early experiment by Cooper and Zubek, “maze-
bright” and “maze-dull” mice were separated on the basis
of their performance in repeatedly running a maze and
then selectively bred for this trait.13 Over several genera-
tions the strains diverged in their maze running abilities
indicating a genetic basis for the differences in learning
and memory involved in this task, and for mice raised in a
“standard” environment heritability of this trait was high.
The researchers then raised the offspring of the genetically
䡬
| |
䡬
Figure 1
Gene-environment interdependence among
maze-bright and maze-dull rats
Source: Meany 2001, 54.
selected lines in two “extreme environments,” a cogni-
tively poor one (dull colors all around, no toys) and a
cognitively enriched one (bright colors and patterns, many
toys). Both the bright and dull lines behaved “stupidly” in
the poor environment and “smartly” in the enriched one,
with the result that the heritability of the trait dropped to
zero in both extreme environments. The bright and dull
mice may well have inherited whatever genes are linked to
intelligence (or maze running aptitude) in mice (which is
what they were bred for), but the heritability of the trait of
intelligence was shown to be highly dependent upon post-
natal environment (see figure 1).14
While the heritability itself of a trait can change signif-
icantly with environmental changes, environment can sig-
nificantly influence traits that are themselves highly
heritable. The high heritability of a trait does not entail that
it is resistant to environmental influences: “High heritability
therefore does not mean low malleability.” 15 An interest-
ing example of this is height. The heritability of height in
most populations is about .9, that is, 90 percent of the
variation in height can be attributed to genetic factors.
The average height of Japanese has increased significantly
since WWII with changes in diet, standard of living, and
health care, while Americans, who were once three inches
taller than their European counterparts, are now sur-
passed by Europeans, with the Dutch on average three
inches taller than Americans.16
Twin Studies
I begin my analysis of the study of AFH with a consider-
ation of the experimental technique—“classical twin
studies”—on the basis of which they make their claims.
The importance of beginning here should require no jus-
tification: Any conclusions derived from empirical research
䡬
stand or fall on the basis of the soundness of the research
methods employed. Inasmuch as I write as a political scien-
tist by training, not a behavioral geneticist (and in this, I
do not differ from AFH), it is my goal in this section to
allow geneticists, as much as possible, to “speak for them-
selves.” I do not claim that anything that I present in this
section is original. Nonetheless, it is vital that those who
have read or heard of the findings of AFH (as well as all
other studies that rely wholly upon the methodology of
twin studies) understand just how little consensus there is
in the scientific community concerning the soundness of
such studies.
Twin studies have been deemed, largely by psycholo-
gists, who most commonly undertake such studies, the
“new workhorse” of behavioral genetics.17 From reading
the conclusions of many such studies purporting to dem-
onstrate the heritability of one or another complex human
trait or system of belief, including those of AFH, one
would assume that twin studies are a scientific procedure
as unproblematic as, say, carbon dating is as a means of
determining the age of organic matter (within an acknowl-
edged range of error). To determine the age of certain
kinds of organic matter, one performs a carbon dating
test; to determine the percentage of a given trait due to
genes as opposed to environment, one performs a “twin
test” (i.e., conducts a twin study). Were this the case, twin
studies would indeed be a research tool of unprecedented 䡬
power in unraveling the etiology of complex traits (or
attitudes or belief systems) and in revealing to what extent
such traits may be said to be the product of genes as
opposed to environment. Such may be the assumption of
some who employ twin studies, but it is a view by no
means uncontested in the scientific community.
In fact, it is not simply the technique of twin studies
that is increasingly questioned by scientists, but many of
the basic assumptions that underlie such studies, for exam-
ple, whether the influences of genes as opposed to envi-
ronment on a given trait can be neatly partitioned into
percentages, or whether viewing nature and nurture as
contrasting entities or causal forces is a paradigm that is
still plausible in light of recent advances in the biological
sciences. Here are several select quotes addressing this issue
from prominent researchers in neurology, genetics and
biobehavioral health:
Nature-Nurture. What a clever phrase. . .but regrettably, in its
nature versus nurture version, it has probably had the negative
effect of facilitating, in some quarters, the distorted perception
of genetic and environmental factors as antagonistic, competi-
tive factors in a simplistic, either-or causal scheme.18
It is not nature or nurture. Nor is it nature and nurture. . . The
search for main effects is a fool’s errand. In the context of mod-
ern molecular biology, it is a quest that is without credibility.
Nature and nurture do not exist in a manner that can ever be
considered independently quantifiable. There is, instead, simply
a continuing process of development that emerges from the
June 2008 | Vol. 6/No. 2 301
| |
Exchange | Genes and Ideologies
constant dialogue between gene and environment . . . Scientific
journals, regrettably, continue to provide examples of research
attempts to quantify the relative contribution of genetic versus
environmental factors to the development of a specific trait, often
intelligence.19
Environment influences the actions of genes, and genes via changes
in the nervous system influence the sensitivity of an organism to
changes in the environment. The two causes are not separable.
Statistical procedures that appear to separate variance according
to genetic and environmental causes do not provide a valid rep-
resentation of physiological reality.20
For example, biologists, geneticists, and medical
researchers commonly emphasize the importance of
gene-environment interaction in the development of the
human brain (including cognitive capacities such as intel-
ligence and language acquisition). To begin with, there
are well known facts concerning the role of postnatal
environment in normal brain development. Most myeli-
nation in the brain is postnatal and synaptogenesis
occurs at a high rate during the first year of life, with
each neuron developing on average 15,000 synapses by
age three.21 Normal brain development does not occur
spontaneously or in a vacuum. It develops in an environ-
mentally dependent maturation process, that is, although
its developmental stages are genetically determined they
must be triggered by specific environmental inputs, known
䡬 as species-specific experiences, within critical periods of
development.22 Language acquisition, for example, is an
environmentally-dependent neurobiological window-of-
opportunity phenomenon. Studies of so-called “feral”
children demonstrate that if children are not exposed
to language it cannot be successfully acquired after
ages 8–11.23 Disruptions of environment-dependent
neurochemical signals during early life may lead to major
abnormalities or deficits in neurodevelopment. Disrup-
tion of critical neurodevelopmental cues can result from
a lack of sensory experience during sensitive periods (e.g.,
lack of sufficient tactile, auditory, and visual stimulation)
or atypical or abnormal patterns of necessary cues due to
extremes of experience (e.g., prolonged or traumatic stress).
Beyond these well-known features of human brain devel-
opment and its environmental dependence, cutting-edge
research on gene/environment interaction presents a sig-
nificant challenge to the cogency of the view that there
could be such a thing as genes for such features of human
cognition as, e.g., intelligence, creativity, or a conservative
political orientation:
Simple, direct mappings between genes and cognitive-level outcomes
are not sustainable. In fact, genes are likely to contribute to much
more general constraints, such as developmental timing, neuro-
nal migration, neuronal type/size/density/orientation, myelina-
tion, lamination, ratio of gray matter to white matter, firing
thresholds, neurotransmitter differences, and so forth . . . More-
over, once one thinks from a truly developmental neuroconstruc-
tivist perspective, it is easy to imagine how even a tiny asynchrony
302 Perspectives on Politics
䡬
or impairment early on in development can have a huge, cascad-
ing impact on the phenotypic outcome.24
It is because of the growing number of claims of this sort
that a recently published book on behavioral genetics, the
result of a project undertaken by The Hastings Center
and the American Association for the Advancement of
Science, notes the following in the Introduction:
Perhaps most surprising, and perhaps in some ways most diffi-
cult for the non-scientist editors of this volume, was the degree
of the disagreement of the facts of the matter that we found
among the scientists . . . While disputes within science are cer-
tainly the norm, not the exception, these disagreements seemed
particularly important. It finally appeared to us that we were in
what might be called a “transformative” moment in behavioral
genetics, in which an awareness of complexity at a variety of
levels is causing a reexamination of many terms and assump-
tions, as well as some hard thinking about whether it will turn
out to be true that identifying statistical correlations between
genotypes and phenotypes will help us to influence or predict
behavior [emphasis in original].25
My aim here is not to examine in any detail this “trans-
formative moment” in behavioral genetics—transformative,
that is, for those who acknowledge that it is even occur-
ring. Rather, as noted, it is to focus on one particular
aspect of this transformative moment that is crucial for an
evaluation of the claims of AFH: questions concerning
the validity of twin studies.
Studies of Twins Raised Together
Classical twin studies compare the similarity of identical
or monozygotic (MZ) twins and fraternal or dizygotic
(DZ) twins in regard to a particular trait of interest (in the
case of AFH, political ideology). MZ twins share 100 per-
cent of their genes while DZ twins share 50 percent of
their genes, on average. The underlying assumption of
twin studies is that greater concordances in a given trait
(or phenotype) among MZ twins, as opposed to DZ twins,
are evidence of a genetic influence or component of the
trait in question. The greater the level of concordance in
MZ as opposed to DZ twins for a given trait, the greater
the genetic component of that trait.26
The conceptual simplicity underlying the methodology
of twin studies requires that a number of empirical pre-
conditions be met, the most basic and important of which
is the Equal Environment Assumption (EEA), i.e., that
MZ twins and DZ twins experience similar environ-
ments. As Hettema, Neale, and Kendler explain, “the valid-
ity of the EEA is crucial because, if incorrect, excess
resemblance of MZ twins compared to DZ twins, usually
ascribed to genetic factors, could be partly or entirely due
to environmental factors.” 27 There is ample evidence that
the EEA, if taken literally, does not hold. For example, in
one review Evans and Martin state that there is overwhelm-
ing empirical evidence to support the claim that MZ twins
are treated more alike than DZ twins. As children, MZ
䡬
| |
䡬
twins are more likely to have the same playmates, share
the same room, and dress alike. As adults, MZ twins are
more likely to keep in contact than same-sex DZ twins.28
Studies have indicated higher rates of interaction among
MZ as opposed to DZ twins; 29 Joseph notes large differ-
ences between MZ and DZ twins in experiences such as
identity confusion (91 percent versus 10 percent), being
brought up as a unit (72 percent versus 19 percent), being
inseparable as children (73 percent versus 19 percent);
and having an extremely strong level of closeness (65 per-
cent versus 19 percent) 30. Studies have also shown that
parents “hold more similar expectations for their MZ than
DZ twins with respect to social responsibility and inde-
pendence.” 31 Furthermore, it is well established that phys-
ical appearance can have a profound effect upon how
individuals are treated.
Differences in physical attractiveness affect the social desirability
judgments which people form of others. . . . From infancy to old
age, there is a strong tendency to attribute more positive quali-
ties to those who are physically attractive relative to those who
are physically unattractive. Attractiveness may be used as a cue to
signal status.32
This fact is particularly significant for twin studies, inas-
much as MZ twins are much more likely to closely resem-
ble one another than DZ twins and are therefore much
more likely to experience the same social treatment that
comes with “social desirability judgments” based upon
physical appearance:
Physically attractive people receive more cultural rewards than
physically unattractive people, . . . a response that develops from
social as well as genetic reasons. Because monozygotic twins have
greater physical resemblance to each other than dizygotic twins,
they could have a greater chance of receiving similar social reac-
tions, which in turn can affect a variety of outcomes.33
Greater environmental similarities by themselves, how-
ever, do not negate the validity of twin studies, since it is
possible that “some environmental exposures that MZ twins
share more often than DZ twins may simply have no
effect on the trait in question.” 34 In fact many, if not most
researchers who support the validity of twin studies accept
that the EEA, if conceived as requiring environments that
are “actually” similar, does not hold.35 Proponents of a
modified, or what is sometimes referred to as trait-
relevant EEA, “recognize that identical twins are treated
more similarly and spend significantly more time together
than fraternals, but claim that the evidence shows that
greater environmental similarity does not lead to higher
concordance [for the specific trait under consideration].” 36
The problem with studies in support of this proposition
is that they have been shown to have significant method-
ological flaws. For example, studies of a tiny subpopulation
of DZ twins mistakenly thought by their parents to be MZ
twins purport to show that the degree of correspondence
between MZ twins still exceeds that of DZ twins.37 Most
䡬
of these studies of “reverse zygosity,” however, relied upon
parental accounts of how they raised their twin children
many years later. Because of problems with biased impres-
sions,38 poor memory, and poor reliability, studies that rely
upon parental recall of their child rearing practices have been
shown to be notoriously unreliable, typically showing reli-
ability measures of only 0.3–0.5.39
Studies of Twins Raised Apart
It has been argued that objections regarding possible con-
founding effects of environment on the results of twin
studies have received their most effective refutation from
so-called studies of twins raised apart. 40 The idea behind
studies of twins raised apart is that inasmuch as the MZ
and DZ twins being studied are not raised in the same
environment, any possible confounding effects that could
result from the possibility that MZ twins are treated more
alike than DZ twins will be eliminated. But studies of
twins raised apart suffer from the same possibility or prob-
ability of significant environmental confounding effects as
studies of twins raised together. Here are five more com-
monly cited possible (or probable) effects (which by no
means comprises a complete list of all of the objections
that have been raised concerning environmental confound-
ing effects in such studies):
1. In no study to date of twins raised apart is it the case 䡬
that all of the twins being studied were separated at
birth; most often, in fact, they have not. In some
well-known studies, twins who have been raised
together up until ages 4, 6, and 11 were held to be
“twins raised apart” for the purposes of the study.41
This astonishing (and oddly unnoticed) fact means
that so-called studies of twins raised apart are in fact
no such thing; at best, they are studies of twins raised
partially apart. This leaves open the possibility of
many years of confounding environmental influ-
ences during the most formative period of develop-
ment in a child’s life.42
2. In no twin study to date is it the case that all of the
twins studied have had no contact after separation.
Often, many of the twins have had ongoing rela-
tionships lasting for many years. The potential con-
founding effects here are significant because given
the demonstrated greater levels of contact between
MZ as opposed to DZ twins—“MZAs [“monozy-
gotic twins raised apart”] are more likely to be aware
of each others’ existence and to have had more pre-
study contact than DZAs [“dizygotic twins raised
apart”]—there is a greater opportunity for MZ twins
to influence each other’s attitudes.43
3. Studies of twins raised apart depend upon the
assumption that the environments in which the sep-
arated twins are raised are “dissimilar.” But over thirty
years ago, researchers found this not to be the case:
June 2008 | Vol. 6/No. 2 303
| |
Exchange | Genes and Ideologies
[Kamin] found that twins who were placed in new homes
by adoption agencies tended to end up in similar envi-
ronments; for example, the adopting families tended to
be from the same social class. Second, in many cases the
separated twins were placed by family members in the
homes of relatives, so that they often went to the same
school, and interacted with each other frequently.44
Because of such problems, in more recent studies
researchers have attempted to “quantify” various
aspects of the home environment into which mem-
bers of the adopted pair were placed, undertaking
such “measures” as counting the number of books
and the availability of household “facilities” such as
power tools and original artwork.45 But the prob-
lem with such (crude) attempts to quantify environ-
mental influences is summed up by Beckwith as
follows:
How do we know what combination of factors both in
the home and in the outside environment may provide
the appropriate mix for development of the complex capa-
bilities or behavioral problems that are being measured?
How can researchers assess the less tangible features of
such environments, such as the daily interactions between
the parent and the child, that cannot be so easily quanti-
fied? While they have taken into account retrospective
impressions of family environment by separated twins,
such retrospective impressions are often thought to be
subject to distortion.46
䡬
It might be argued that none of this constitutes a
problem unless it could be shown that separated
MZ twins are more likely to be placed in more sim-
ilar homes than DZ twins, which might be consid-
ered unlikely,47 but in the absence of knowledge as
to what the relevant environmental “similarities”
might be, and whether or not they could affect the
trait in question, it is impossible to judge the cor-
rectness of this assertion. But beyond this, Kamin
and Goldberger note that in one of the well-known
studies of twins raised apart, the investigators sub-
mitted results of two questionnaires dealing with
childhood family environment to a factor analysis.
For one factor, labeled Support, the average MZA
and DZA r was 0.41 and −0.01, respectively. These
results might be interpreted as indicating that selec-
tive placement was substantially greater for MZAs
than for DZAs, leading to more similarity of out-
comes for the MZAs.48
4. In all studies of twins raised apart, the potential
confounding effects of the much greater physical
resemblance of MZ twins remains a significant prob-
lem: “Aside from the environment in which they are
raised, the identical appearance of [MZ] twins, even
though they are separated, can mean that their inter-
actions with the outside world are much more sim-
ilar than for any two randomly selected people [and
for any randomly selected pair of DZ twins].” 49
304 Perspectives on Politics
䡬
5. Studies of twins raised apart are particularly vulner-
able to selection bias. In many studies, twins “were
self-recruited, attracted by reports of reunited twins
appearing in the press.” 50 Inasmuch as such stories
were full of accounts of the uncanny similarities
among MZ twins, there is the potential that those
who either volunteered for, or agreed to participate
in such studies, did so out of a pre-existing belief in
their own “uncanny similarities.” 51
I have not, of course, demonstrated that any of the
confounding effects I have noted above are responsible for
the concordances observed by Alford, Funk, and Hibbing
which they deem to be statistically significant—in the case
of AFH, the confounding effects are those associated with
studies of twins raised together—but neither have AFH
proved that genes rather than confounding effects are respon-
sible. As with most twin studies, all that AFH have observed
are a series of concordances in the self-reported attitudes
of twins which they deem to be statistically significant.
On the basis of these observed concordances, they then
hypothesize that the cause of these concordances is genes. They
have not discovered a set of genes corresponding to their
hypothesis of genetic causation (nor have any other
researchers in regard to complex cognitive and psycholog-
ical traits). What I am hypothesizing—or presenting as a
counter-hypothesis—is that the cause of these concordances
is one or more of the environmental concordances noted above,
a causal explanation that seems a good deal more obvious
and intuitive than one which invokes genes.
I cannot, of course, employ what I have argued is a
fundamentally flawed methodology to prove that the con-
cordances observed by AFH are in fact due to environ-
ment as opposed to genes (i.e., are due to what amount to
confounding effects from the perspective of the method-
ology).52 It might appear at this point that we have two
equally valid explanations for a phenomenon and that,
embracing the principle of lex parsimonae, we should
embrace the less complicated formulation. But we need
not go so far as to ask which explanation is less compli-
cated, for the principle of lex parsimonae depends upon an
assumption that both explanations are equally valid.53 As
I shall argue in what follows, such is not the case.
The Findings of Alford, Funk,
and Hibbing
In my opinion, the problems with classical twin studies,
if applied to as complex a system of belief as a political
ideology, are overwhelming.54 But now I would like to
proceed as if there were no such problems, and turn to
a consideration of the results of the twin study of AFH.
AFH essentially replicate (with a different study popula-
tion) an earlier study by Bouchard et al., in which
separated MZ and DZ twins were administered the
Wilson-Patterson Attitude Inventory (W-P), which is
䡬
| |
䡬
intended to provide an “overall index of conservatism.” 55
In administering the inventory, subjects are presented
with a word or phrase, such as “pajama parties” or “com-
puter music,” to which respondents must select from the
responses “yes,” “no,” and “uncertain.” AFH selected 28
items from the 50-item W-P inventory that they consid-
ered to be more politically relevant.56 From these responses
the authors worked out estimates of the heritability of
“political ideology” (in this case, conservatism) by com-
paring the responses for MZ pairs with those of DZ
pairs. The conclusion the authors draw from the compar-
ison of MZ and DZ twin responses is that “political
attitudes are influenced much more by genetics than by
parental socialization.” 57
In characterizing the data from W-P inventory the
authors state that “we see from [the table of responses to
the W-P items] that the impact of shared environment
exceeds that of heritability in only four of the twenty eight
items and the mean estimate of heritability for the 28
W-P items is .32, compared to a mean estimate of shared
environmental influence of .16.” 58 They further com-
ment that “while the individual items provide interesting
variations, the purpose of the W-P is to provide an overall
index of conservatism.” 59 To this end, they compute a
simple index by assigning a value of ⫹1 to any “conserva-
tive” response, like a “no” on modern art, and a ⫺1 to any
“liberal” response, like a “yes” on modern art. Hence, in
calculating conservatism, attitudes toward astrology, mod-
ern art, and military drills are afforded precisely the same
weight as attitudes toward abortion and gay rights. Fur-
thermore, an examination of the table of “Attitude Items”
from the W-P scale reveals that the four items for which,
according to the authors’ estimates, environment exceeds
heritability, are attitudes toward abortion, gay rights, liber-
als, and living together (see table 1).
It is not entirely clear what the authors intend by the
term “conservative,” nor does the term, taken by itself,
have a single and unambiguous denotation or connota-
tion (a point I shall return to). For the moment and for
simplicity’s sake, I will identify the term “conservative”
with those distinct political views held by voters at the
present time in the United States 60 who are self-identified as
“social-conservatives” (as opposed to, e.g., “fiscal conser-
vatives”); this includes such groups as “right” or “conser-
vative” Republicans and “Christian conservatives.” Among
groups such as these, perhaps no issues are held to be of
greater importance at the present time in the United States,
or more sharply divide conservatives from liberals, than
their views concerning abortion and gay rights. For exam-
ple, consider the results of a Zogby poll of voters con-
ducted just before the last presidential election (table 2).
In response to the question “Which of the following moral
issues most influenced your vote in this election?,” 32.8
percent of conservative voters ranked the Iraq war first,
followed by abortion (21.4 percent) and same sex mar-
䡬
riage (18.7 percent); and in response to the question,
“Which is the most urgent moral problem in American
culture?,” the largest percentage of conservatives (30 per-
cent) responded that it was abortion, followed by same
sex-marriage (26.5 percent). Recent debates over the views
of Supreme Court nominees on abortion, along with the
new push among conservatives for a “marriage amend-
ment” to the Constitution, demonstrate that abortion and
gay rights are the central concerns of conservatives which
divide them from liberals at the present time in the United
States.
According to AFH, “it has long been known that cer-
tain issues seem ‘hard’ to people, while others seem ‘easy,’
presumably because some issues trigger ‘gut responses’
while others do not,” and they conjecture that “in light
of the new findings, one distinct possibility is that easy
‘gut’ issues tend to be those that are more heritable.” 61 If
by easy the authors intend something like “obvious” or
“unambiguous” or clear to those who hold them (as
opposed to “easy to resolve politically”), then it is indeed
true that persons often feel most impassioned—as in hav-
ing a “gut response”—about those issues that they con-
sider both to be critically important and to be, from
their perspective, unambiguously right or wrong. I can
think of no two issues at the present time in the United
States that more accurately fit this description of a “gut
issue” for conservatives (more heritable as AFH conjec- 䡬
ture) than abortion and gay rights (less heritable accord-
ing to their data).
To be sure, there are other items on the edited W-P
index presented by AFH that one would tend to correlate
with contemporary social conservatism in the United States,
the most important of which at the present time is, per-
haps, school prayer. But even an issue such as school prayer
trails behind abortion and gay rights in terms of impor-
tance to conservatives. Opposition to abortion and gay
rights is so central and defining for social conservatives at
present in the United States that it is not entirely clear
that one could be a social conservative and agree with
either of these. And this is even more obviously the case
concerning whether or not one “agrees with liberals,”
another attitude item from the W-P index found not to be
heritable by AFH. Liberals are, in effect, the “antithesis”
of conservatives, an opposition exploited by AFH them-
selves in working out their dichotomous liberal-conservative
phenotypes/genotypes. Could one ever conceivably be a
conservative and simultaneously agree with liberals?
The problem with assigning an equal numeric value
(⫹1, ⫺1) to all of the attitude items on the edited W-P
index should now be apparent. If we are using an additive
model in which overall conservatism is calculated by add-
ing up one’s plus (⫹1) or minus (⫺1) responses to the
items on the W-P index, such an equal numerical weight-
ing allows for the following anomalous result: One could
agree with abortion rights, agree with gay rights, disagree
June 2008 | Vol. 6/No. 2 305
| |
䡬

Exchange | Genes and Ideologies

Table 1
Genetic and environmental influences on political attitudes: The 28 individual
Wilson-Patterson Items
Polychoric Correlation
Shared Unshared Z for
MZ DZ
Heritability Environment Environment (MZ − DZ)
Attitude Item Corr. n Corr. n 2 * (MZ − DZ) (2 * DZ) − MZ 1 − MZ Difference*
School prayer 0.66 2,687 0.46 1,774 0.41 0.25 0.34 9.83
Property tax 0.47 2,643 0.46 1,774 0.41 0.06 0.53 7.66
Moral Majority 0.42 2,614 0.22 1,717 0.40 0.03 0.58 7.16
Capitalism 0.53 2,609 0.34 1,720 0.39 0.14 0.47 7.85
Astrology 0.48 2,631 0.28 1,721 0.39 0.09 0.52 7.39
The draft 0.41 2,641 0.21 1,753 0.38 0.02 0.59 6.94
Pacifism 0.34 2,576 0.15 1,686 0.38 −0.04 0.66 6.43
Unions 0.44 2,661 0.26 1,752 0.37 0.07 0.56 6.89
Republicans 0.48 2,627 0.30 1,734 0.36 0.12 0.52 6.91
Socialism 0.43 2,616 0.25 1,726 0.36 0.07 0.57 6.53
Foreign aid 0.41 2,669 0.23 1,771 0.35 0.06 0.59 6.42
X-rated movies 0.63 2,685 0.46 1,783 0.35 0.28 0.37 8.15
Immigration 0.45 2,658 0.29 1,784 0.33 0.12 0.55 6.20
Women’s liberation 0.46 2,666 0.30 1,779 0.33 0.13 0.54 6.27
Death penalty 0.56 2,684 0.40 1,775 0.32 0.24 0.44 6.83
Censorship 0.40 2,629 0.25 1,718 0.30 0.10 0.60 5.36
Living together 0.67 2,679 0.52 1,771 0.30 0.37 0.33 7.54
Military drill 0.38 2,625 0.24 1,733 0.29 0.09 0.62 5.24
Gay rights 0.60 2,658 0.46 1,767 0.28 0.32 0.40 6.26
Segregation 0.38 2,653 0.24 1,743 0.27 0.11 0.62 4.83
Busing 0.43 2,670 0.30 1,766 0.26 0.16 0.57 4.92
Nuclear power 0.42 2,646 0.29 1,744 0.26 0.16 0.58 4.84
䡬 Democrats 0.47 2,639 0.34 1,726 0.26 0.21 0.53 4.96
Divorce 0.47 2,659 0.34 1,765 0.26 0.21 0.53 4.99
Abortion 0.64 2,668 0.52 1,768 0.25 0.39 0.36 6.23
Modern art 0.43 2,662 0.30 1,765 0.25 0.18 0.57 4.69
Federal housing 0.36 2,665 0.26 1,766 0.20 0.16 0.64 3.61
Liberals 0.44 2,629 0.35 1,734 0.18 0.26 0.56 3.40
28-item mean 0.47 2,648 0.31 1,748 0.32 0.16 0.53
Source: Access to the data provided by Eaves et al., principal investigators, Virginia 30k twin study.
*The MZ-DZ correlation difference is statistically significant for all the table items at the 0.01 level or above.

with prayer in public schools, disagree with the moral
majority, agree with women’s liberation and in general,
agree with liberals, and still be considered a conservative
because one’s responses on the remaining items would still
add up to a number greater than the sum of one’s “non-
conservative” responses. Conservatism (at the present time
in the United States) cannot accurately be measured by
assigning an equal numeric value to one’s attitudes con-
cerning gay rights, abortion, school prayer, and for liber-
als, one’s attitudes toward astrology, modern art, federal
housing, and military drills.
It might be objected that none of this matters because
according to AFH all of the items on the W-P index
showed, as they assert, a “statistically significant” genetic
component.62 But the problem with this objection is that
what the authors are claiming to have demonstrated is
that genes count for more than environment when it comes
to the transmission of political ideologies. Their claims do
not concern the heritability of “discrete” and potentially
disconnected attitudes of political relevance. To the extent
that their claim does concern the heritability of political
ideologies, it is not obviously supported by their own data.
Liberal and Conservative
“Phenotypes”
Central to evaluating the claim that political ideologies
are heritable, or have a heritable component, is an under-
standing of precisely what is intended by such a claim.
What Alford, Funk and Hibbing say in this regard is nei-
ther entirely clear nor consistent. For example, they assert
that “the connection between genes and attitudes may not
involve specific attitudes as much as the flexibility of atti-
tudes” 63 and, “individual genes for individual behaviors

306 Perspectives on Politics

䡬
| |
䡬

Table 2
Zogby Poll, 2004 presidential election
All voters Liberals Moderates Conservatives?
Which of the following most influenced your vote in this election?
Iraq War 42.3% 54.6% 47.2% 32.8%
Abortion 12.8% 3.3% 6.4% 21.4%
Same-Sex Marriage 9.3% 2.8% 4.7% 18.7%
Poverty 7.4% 16.6% 7.8% 0.9%
Health Care 5.6% 9.1% 9.0% 1.3%
Preventing Stem-Cell Research 2.1% 1.5% 3.1% 2.0%
None 7.0% 2.4% 12.0% 9.0%
Other 13.3% 1.9% 9.1% 13.3%
Which of the following is the most urgent moral problem in American culture?
Greed/Materialism 32.6% 34.9% 40.8% 24.3%
Poverty/Economic Justice 30.6% 61.4% 37.6% 6.0%
Abortion 15.7% 0.3% 5.8% 30.0%
Same-Sex Marriage 12.3% 1.5% 6.7% 26.5%
None/No Response 8.9% 1.9% 9.1% 13.3%
Nat National survey of 10,660 who voted in last week’s presidential election. Survey was conducted Nov. 3–Nov. 4. Sample drawn
from voters of all parties. Margin of error 1 percentage point.
Source: Zogby International.

do not exist.” 64 Yet at the same time, they set forth the
features of what they term liberal and conservative phe-
notypes (corresponding to liberal and conservative geno-
types), which are largely a collection specific of attitudes; for
example, a favorable attitude toward unbending moral
and behavioral codes and swift and severe punishments
for violations of that code (conservative phenotype), or an
unfavorable view of preset punishment, hierarchy, and social
inequality (liberal phenotype). They also propose specific
genes for these specific attitudes.65 I shall examine each of
what appear to be two distinct claims or proposals in turn—
specific attitudes are not heritable, but rather a “flexibility
of attitudes,” and specific attitudes are heritable, namely
the specific attitudes that comprise the liberal-conservative
phenotypes. I begin with the latter.
In positing the existence of liberal and conservative “phe-
notypes” (and genotypes), AFH claim to have answered
the following question:
The left/right political divide, in the sense AFH intend,
is a decidedly modern phenomenon, with distinct histor-
ical origins in the eighteenth century. Originally, the defin- 䡬
ing point on the ideological spectrum was attitudes toward
the ancien regime in France; the right implied support for
aristocratic, royal, or clerical interests, while the left implied
radical and egalitarian sympathies, as well as support for
laissez-faire capitalism and free markets.67 Liberalism, in
an identifiably modern form, has its origins in the seven-
teenth and eighteenth centuries in England, France, and
America, while conservatism, as a self-conscious political
ideology defined, in part, by its opposition to Enlighten-
ment liberal principles, was born in the aftermath of the
French Revolution.68
Nor is it the case that the “package of attitudes” held by
liberals and conservatives today is remarkably similar to
those held in the seventeenth and eighteenth centuries.
Consider conservatism:

Why is a reasonably standard left-right spectrum widely appli-
cable cross-culturally and over time? The universal left-right ele-
ments of belief systems around the world and over decades are
difficult to explain . . . [T]he package of attitudes held, for exam-
ple, by conservatives in the United States is remarkably similar to
that held by conservatives in other cultures and at earlier times in
American history.66
But if this is the problem that the authors claim to have
solved, then they have solved a non-existent problem, for
their characterization of a standard left-right spectrum as
universal, existing in all cultures in all times, is completely
historically inaccurate.
Modern American conservatism is very different from European
conservatism, or from conservatism traditionally understood. For
one thing, conservatism in this country is “modern,” and for
another, it is “American.” Ours is not the “throne and altar”
conservatism that once defined European conservatism, and that
is still characteristic of many Europeans on the right. These con-
servatives were true reactionaries. They sought to preserve the
ancien regime and the prerogatives of king and church against
the arrival of modern science, modern capitalism, and modern
democracy.69
Modern liberalism differs in fundamental respects from
what is sometimes called classical liberalism. For example,
classical liberals, associated with early modern figures such

June 2008 | Vol. 6/No. 2 307

䡬
| |
Exchange | Genes and Ideologies
as Locke and Smith were much more tolerant of eco-
nomic inequality. By contrast, by the end of the nine-
teenth and early twentieth century, a growing body of
liberal thought held that the maintenance of “liberal free-
dom” required a certain amount of government interven-
tion in the economy and the creation of social welfare
programs of one form or another 70 (an idea turned into
partial political reality in the United States during the
Roosevelt administration 71 ). It is this latter conception of
liberalism—“egalitarian” or “welfare state” liberalism—
given canonical formulation in the late twentieth century
by John Rawls’s in A Theory of Justice—that today is com-
monly associated with liberalism in contemporary Amer-
ican society.
Not only is it not the case that the “package of atti-
tudes” associated with liberalism and conservatism, left
and right, have remained remarkably stable for the past
300 years, it is also not the case that there has ever existed
any single “package of attitudes” that define liberalism
and conservatism and unite all of their varied historical
manifestations. It is doubtful that any ideology possesses
an unchanging “core cluster” of political beliefs that effec-
tively distinguishes it from all other ideologies, which makes
the assumption of AFH that there exist discrete liberal
and conservative “phenotypes” particularly problematic.
As evolving and shifting historical modes of thought, ide-
䡬 ologies are fundamentally hybrid and interconnecting phe-
nomena. As Jeremy Waldron notes:
It is fruitless, not only to look for a core of common characteris-
tics, but also to think that we can find distinguishing or peculiar
characteristics which differentiate views in one tradition from
those in another. Liberal moderatism fades into conservatism;
the conservative’s concern for community matches the socialist’s;
the socialist claims to take the liberal concern for freedom more
seriously than liberals themselves, and so on.72
In a similar vein, Alan Ryan writes:
Anyone trying to give a brief account of liberalism is immedi-
ately faced with an embarrassing question: are we dealing with
liberalism or with liberalisms? It is easy to list famous liberals; it
is harder to say what they have in common . . . They do not
agree about the boundaries of toleration, the legitimacy of the
welfare state, and the virtues of democracy, to take three rather
central political issues. They do not even agree on the nature of
the liberty they think liberals ought to seek (Berlin 1969,
122–34).73
The problem with attempting to propose a core cluster
of attitudes that distinguishes, in all times and places, one
political ideology from another, is evident in AFH’s attempt
to characterize the liberal and conservative phenotypes,
which they do as follows 74 :
Conservative phenotype: Pro-attitudes toward in-group unity, strong
leadership, clear and unbending moral and behavioral codes,
swift and severe punishments for violations of this code, and
“systematization;” a suspicion of out-groups; toleration of social
inequality; a pessimistic view of human nature.
308 Perspectives on Politics
䡬
Liberal phenotype: Pro-attitudes toward out-groups (greater tol-
erance), context-dependent rather than rule-based approaches
to proper behavior, and group togetherness. Con-attitudes toward
preset punishments, hierarchy, certainty, strong leadership, and
inequality; an optimistic view of human nature; greater
empathy.75
Not surprisingly, in light of what has just been noted
concerning the historical and hybrid nature of political
ideologies, much of what AFH classify as distinctive of
liberal or conservative attitudes (or phenotypes) could eas-
ily characterize either. For example, a “faith in human
nature” is characterized as a liberal belief (or attitude), but
some of our most basic liberal institutions have been
defended on the basis of a thoroughly pessimistic view of
human nature, e.g., Madison in the Federalist Papers
famously wrote that “the latent causes of faction are sown
in the seeds of human nature,” and “have rendered per-
sons much more disposed to vex and oppress each other
than to cooperate for the common good.” 76 A desire for
“clear and unbending moral and behavioral codes,” clas-
sified as a conservative attitude, could just as well charac-
terize a liberal supporter of constitutionalism and the rule
of law (as opposed to say, the whims of a king, pope, or
mullah), or a liberal advocate of an expansive list of inter-
national human rights. In fact, liberals from Kant to Rawls
have been accused of being “rule worshippers” in their
insistence that certain moral principles, such as “rights,”
be upheld regardless of the consequences,77 which does
not square well with the sweeping characterization of
liberals as having a “context-dependent” rather than “rule-
dependent” approach to human nature. Taking a context-
dependent approach to human nature is, if anything, much
more characteristic of conservatism in its varied historical
manifestations, inasmuch as one of the defining features
of conservatism is a belief that political wisdom is embod-
ied in the particular “context-dependent” beliefs and prac-
tices of a particular society rather than in “abstract universal
principles.” 78 Liberals may or may not be averse to inequal-
ity depending upon what type of inequality is being referred
to and what type of liberal one considers oneself. As noted,
classical liberals, represented in recent decades by such
thinkers as Hyeck, Freidman, and Nozick (who are some-
times called libertarians—and even “conservatives”!) have
no problem accepting large inequalities in wealth and
income and are generally opposed to redistributive taxa-
tion. And in any given individual today, one might find a
mix of what are listed by AFH as liberal and conservative
attitudes. Conservative Catholics, for example, tend to
support clear and unbending moral and behavioral codes
(conservative phenotype), but are also opposed to capital
punishment (liberal phenotype); and while they accept
the hierarchy of the Church (conservative phenotype), they
oppose large-scale inequalities of wealth and income and
historically have been champions of the poor (liberal
phenotype).
䡬
| |
䡬
Liberalism and conservatism cannot be captured by a
distinct and unchanging core “cluster of political atti-
tudes.” Any attempt to do so is a distortion of the com-
plex history of these two ideologies. Nor is this problem
alleviated if we shift to the more general—and more
ambiguous—terms left and right (as when AFH speak of
a “left-right divide”). The idea might be that left and right
are political “spectra” or “axes” along which all ideologies
can be ranked. But the problem here is the same as that
with liberalism and conservatism: There exist no clear cut,
ahistorical clusters of attitudes that can comprise the end-
points, as it were, of the “left-right spectrum,” a spectrum
along which we could meaningfully place all of the ideol-
ogies that have ever existed. AFH write, in fact, as if left
and right are not really a spectrum at all, but rather are
terms equivalent to liberalism and conservatism. Regard-
less of what one thinks of the usefulness of the left/right
axis as a classificatory scheme, it is customary to charac-
terize ideologies as left of liberalism (e.g., socialism and
communism) and right of conservatism (e.g., reaction-
ism). And the usefulness of such a linear left-right spec-
trum is increasingly challenged.79 There is no real consensus
as to what exactly this spectrum is supposed to be mea-
suring and where precisely to place specific ideologies.
Talk of ideologies and a political spectrum points to
another significant problem with AFH’s characterization
of liberalism and conservatism; for they write as if these
two ideologies divide the political universe between them,
as if they are the only political ideologies that have ever
existed. Far from dividing the political universe, liberal-
ism and conservatism have been pre-dated by, had to share
the historical stage with, and had to compete with ideol-
ogies as diverse as theocratism, monarchism, anarchism,
Marxism, Stalinism, Maoism, socialism, fascism, and total-
itarianism. And what of non-Western political ideologies
throughout history? Ancient Egypt, Buddhism, Confu-
cianism, Hinduism, and Islam have all given rise to man-
ifold and distinctive political ideologies throughout history,
all of which could be called liberal or conservative, left or
right, only by engaging in extreme procrustean distortion.
In characterizing the liberal-conservative divide as in
effect “universal,” AFH have committed the fallacy of
taking the local and particular for the universal. In char-
acterizing the political attitudes of twenty-first century
Americans they assume that they have characterized the
political attitudes of all mankind in all times and places
(or mankind abstracted from all times and places). The
commission of such a fallacy is possible only by paying
insufficient attention to both history and culture, and
the lack of such attention seems naturally to be fostered
by two of AFH’s assumptions. The first, already exam-
ined, is that there exist trans-cultural and trans-historical
liberal and conservative phenotypes/genotypes. The sec-
ond is that “political attitudes are influenced much more
by genetics than by parental socialization.” 80 For “paren-
䡬
tal socialization” can only be socialization within a par-
ticular culture in a particular historical period, and if
such socialization matters less than genes in the etiology
of political attitudes, then the same is true of culture and
history.81
Yet even if, for the sake of argument, we could agree
on a universal set of attitudes that meaningfully distin-
guishes liberal from conservative ideologies, or the left
from the right in all times and places, there remains the
difficulty that while AFH acknowledge that heritability is
not a matter of the genetic transmission of particular
attitudes, we are presented with “phenotypes” that are
essentially a collection of particular attitudes.82 In fact,
the authors go so far as to tie specific attitudes to specific
genes (or groups of genes), conjecturing that while the
genes that make up the liberal and conservative pheno-
types tend to “move together,” they need not always do
so:
Even if the individual genes involved with absolutism or contex-
tualism [the broader “world orientation” phenotypes of which,
according to AFH, the liberal-conservative phenotypes are a part]
tend to move together, this does not mean that they always do.
Some individuals may carry, say, the absolutist’s aversion to out-
groups but a contextualist’s rejection of a universal behavioral
code.83
This characterization as to how genes—and attitudes—
might move warrants comment. It suggests what might be 䡬
termed a discrete “attitude unit” view of political ideolo-
gies. That is, an ideology is composed of a set of distinct
attitudes or beliefs, each particular attitude the product of
a corresponding gene (or set of genes). These specific
genes—and the corresponding attitudes—can, even if they
tend to be grouped in certain patterns (as in the liberal-
conservative phenotypes), be rearranged in conceivably
any pattern (depending on how the discrete genes move).
Were this true, ideologies would lack any overarching (or
underlying) “themes” or principles that united the beliefs
of which they were comprised in some coherent fashion.
Ideologies would in effect be a random concatenation of
discrete “attitude units.”
Now AFH do say that “for the most part” the beliefs
that comprise the liberal-conservative phenotypes do
“move together.” 84 But why do they “move together”?
For example, why are libertarians committed to a strong
conception of equal (negative) rights and simultaneously
accepting of inequalities in wealth and income? Were we
to assume that these two sets of attitudes—support for
certain basic equal rights, acceptability of unequal wealth
and income—as discrete attitude units, just happened to
move together, then libertarianism, and every other ide-
ology, would ultimately be meaningless, a happenstance
concatenation of discrete attitudes.85 In libertarianism,
these two attitudes are linked as part of a normative sys-
tem; to understand libertarianism is to understand the
reason why these two beliefs cohere. To say that they
June 2008 | Vol. 6/No. 2 309
| |
Exchange | Genes and Ideologies
cohere because that’s just the way that certain genes tend
to move (as part of the libertarian phenotype) is to fail to
understand the ideology; or put another way, it is to fail
to comprehend the ideology as an ideology. Furthermore,
could genes move in such a way that an individual ends
up holding, for example, mutually exclusive proposi-
tions, such as a belief in radical equality and a favorable
view of hierarchy? What is to preclude such a possibility
on a view in which simply the movement of genes deter-
mines which group of beliefs will end up being lumped
together to form an ideology? (And if one is inclined to
deny that this could ever happen, one should ask oneself
if this denial is based upon principles of genetics, or
rather, principles of reason and rationality).
To return to the problem of the genetic transmission of
particular political attitudes, what might be the problem
with such an assumption? An attitude such as a favorable
view of swift and severe punishments presupposes an enor-
mous amount of latent knowledge somehow encoded in
our genes—of the existence and characteristics of regula-
tive social norms (e.g., laws, conventions, taboos), of the
existence and characteristics of the social practice of pun-
ishment for violations of those norms, of the normative
significance of both social norms and punishment, of the
normative significance of such punishments being swift
and severe (as opposed to being slow and lenient). In fact,
䡬 it presupposes something like an entire culture. Punish-
ment is a normative social practice, an institutional means
of maintaining social order and control and, as such, is
what John Searle calls an institutional fact, like other social
institutions and practices such as money, property, and
marriage.86 Institutional facts determine socially defined
behaviors, beliefs, and attitudes, and in the words of Sir
Michael Rutter, “genes do not, and cannot, code for socially
defined behaviors.” 87
Personality and Political Ideology
One way to avoid these difficulties is to assume that AFH
do not actually mean that specific political attitudes are
heritable (despite all that they say).
This brings us to their suggestion that “the connection
between genes and attitudes may not involve specific atti-
tudes as much as the flexibility of attitudes . . . Individual
genes for individual behaviors do not exist.” It is for this
reason, presumably, that AFH invoke the concept of “per-
sonality” in explaining the heritability of political ideolo-
gies: “We . . . predict that attitudes on political issues
tracking most closely to central personality traits are gen-
erally heritable since personality traits are generally heri-
table and since the heritability of social attitudes is likely
derivative of the heritability of various personality traits.” 88
The claim appears to be that personality traits are her-
itable, not political ideologies or political attitudes, and
that these heritable personality traits are correlated with
310 Perspectives on Politics
䡬
political beliefs, i.e., political ideologies are “derivative of ”
heritable personality traits. The idea that it is personality
traits and not political ideologies that are heritable might
appear to answer some of the objections raised above. We
might conjecture that a given heritable personality trait
would manifest itself in terms of one political ideology in
one culture at a given time and in another political ideol-
ogy in another culture or the same culture at a different
time. Thus we would say that genetically transmitted per-
sonality trait X, while correlated with political attitude(s)
Z among a specific population, is not itself attitude Z, but
in a different, time, place, and culture, would manifest
itself as attitude W.
The idea that there is some connection between spe-
cific personality traits and the holding of specific political
ideologies has a long and venerable pedigree in political
science and political psychology, extending back at least as
far as Plato’s Republic. It is also an idea that does not in
any way depend upon assumptions concerning the heri-
tability of personality traits, i.e., personality traits have
often been linked by theorists with political ideologies in
the absence of any distinct assumptions as to how those
personality traits are acquired (e.g., genes, early childhood
experiences, education, parental/societal indoctrination,
etc.).
In making causal connections between so-called per-
sonality traits and particular beliefs or attitudes there is a
constant temptation to define a given trait as simply hav-
ing the particular set of attitudes for which one is trying to
account. But these personality traits, whatever they may
be, cannot be defined in circular fashion by the very atti-
tudes to which they are supposed to give rise; we must be
able to identify a distinct set of traits that while correlated
with a distinct set of beliefs can be characterized apart
from those beliefs. As noted earlier, AFH talk of specific
genes (or a specific gene) as being associated with specific
political beliefs in a quite precise way.89 If we are to assume
that it is personality traits that are heritable then it would
be necessary to correlate a distinct personality trait with
each of the pro and con attitudes the authors list as com-
prising the liberal-conservative phenotypes, keeping in
mind that the personality trait in question cannot be
defined as, e.g., “an aversion to out groups” (conservative
phenotype) or “a dislike of universal behavioral codes”
(liberal phenotype).
The authors do not attempt this, and it is doubtful that
one could ever imagine a distinct personality trait to be
correlated with every one of the attitudes listed as com-
prising the liberal-conservative phenotypes. And any
attempt to do so would be extremely problematic, if not
silly. What specific (and presumably heritable) personality
trait would one propose to correlate with “a desire to take
a more context dependent rather than rule-dependent
approach to proper behavior” (liberal phenotype) or “a
fondness for systematization” (conservative phenotype)?
䡬
| |
䡬
And any personality trait proposed would entail that lib-
eral and conservative individuals would have or lack the
personality trait in question depending upon whether or
not they were liberals or conservatives. Thus, were one to
suggest that, e.g., “irascibility”—a tendency to be easily
angered—was the heritable personality trait correlated with
“a desire for swift and severe punishments,” this would
entail that there could be no (or very few) “irascible” liberals.
The one suggestion that AFH do give concerning those
heritable personality traits that are presumed to give rise
to distinct political ideologies is a reference to “openness”:
“One of psychology’s ‘Big 5’ personality traits is general
openness and it seems likely degree of openness is relevant
to the political arena. Liberals and conservatives, on aver-
age, differ in their openness to atheism, homosexuality,
communism, immigration, and counter-cultural activi-
ties.” 90 Of course, conservatives could be characterized as
being more “open” than liberals to, e.g., a theistic world-
view, maintaining traditional moral standards concerning
sexuality, the institution of private property, immigration
quotas, and upholding traditional cultural practices in gen-
eral. Unless one assumes that one can only be “open” to
liberal principles, we need a more informative definition
of “openness” to evaluate the authors’ claim. And I will
simply assert that it is doubtful that any non-tendentious
use of the term “openness” as a personality trait will yield
the result that only liberals can have an “open” personality.91
Conclusion
I have argued the following: First, that the research tech-
nique employed by AFH—twin studies—is, according
to many prominent scientists, based upon a faulty para-
digm, namely, that in studying the origin of complex
traits or attitudes or beliefs, genes and environment (or
nature and nurture) can be meaningfully partitioned in
such a way that the causal effects of each can be precisely
quantified by statistical methods. Second, that the same
research technique is, according to many prominent sci-
entists, susceptible to so many confounding effects that
there are good reasons to be skeptical of claims of heri-
tability based solely upon such studies, particularly in
relation to attitudes that are critically dependent upon
environment (socialization and culture) for their content
and specificity. Third, the defects of twin studies notwith-
standing, the results of the study of AFH do not obvi-
ously support the conclusions they draw from them.
Fourth, the assumption that there could be such a thing
as liberal and conservative “phenotypes” depends upon
historically incorrect assumptions concerning both the
universality and ahistorical nature of liberalism and con-
servatism, and the mistaken view that liberalism and
conservatism can each be reduced to a distinct set of
unchanging “core attitudes.” Fifth, to the extent that what
the authors are really proposing is genetically transmitted
personality traits rather than distinct political attitudes,
䡬
it is highly problematic to assume that there is a distinct
personality trait for each attitude that the authors iden-
tify as comprising the liberal and conservative phenotypes.
I would like to conclude with the following consider-
ation regarding the value of an explanatory hypothesis
(for I consider what AFH have presented—for all of the
reasons noted above—to be not a scientific proof, but an
explanatory hypothesis). One way to evaluate the sound-
ness of a hypothesis is to ask whether or not it can explain
the phenomenon under consideration better than other
explanations. In the present case, the assumption that polit-
ical ideologies are genetically transmitted, rather than
explaining the phenomena better than, say, traditional his-
torical and cultural and sociological explanations, render
them mysterious, if not incomprehensible. For example,
the historical advent, development, and many transforma-
tions of liberalism and conservatism as political ideologies
are well (but by no means completely) understood as his-
torical phenomenon. If genes count for more than envi-
ronment in explaining the phenomena of liberalism and
conservatism, then these phenomena become utterly
incomprehensible. Why should liberalism, as a “pheno-
type,” manifest itself in Europe in the seventeenth, eigh-
teenth and nineteenth centuries? Why should conservatism,
as a phenotype, manifest itself (at least in its most well-
known form) in the wake of the French Revolution? Why
should great liberal and conservative thinkers—Locke, 䡬
Smith, Burke, de Maistre—suddenly appear on the world
stage when and where they do? How to explain the differ-
ences between classical liberalism and its modern mani-
festations, and how one developed out of the other? How
to explain the transformation of conservatism from its
earlier “reactionary” forms to its modern incarnations? How
to explain the birth, spread, and eventual demise of com-
munism, one of the most important political ideologies of
the twentieth century (and which at one time was viewed
as the greatest ideological threat to both liberalism and
conservatism)? What of fascism? What of anarchism? What
of the resurgence of nationalism in Eastern Europe with
the collapse of the Soviet Union? And so on and so forth.
What possible “genetic story” could account for these undis-
puted historical phenomena?
But it is not only history that is rendered incomprehen-
sible by the assumption that political ideologies are genet-
ically transmitted (and that the ideological universe is
divided between universal and unchanging “left” and
“right” “phenotypes”). A good deal of public opinion data
becomes equally incomprehensible. For example, if one
looks at conservative self-identification from 1972–2004,
one finds some obvious correlations. Income shows a high
and persistent correlation with self-identification as a con-
servative: As of 2004, 18 percent of those in the 0–16
income percentile self identified as conservatives, com-
pared with 48 percent in the 96–100 percentile. In addi-
tion, race shows a high correlation with self-identification
June 2008 | Vol. 6/No. 2 311
| |
Exchange | Genes and Ideologies
as a conservative: 36 percent of whites as opposed to 15
percent of blacks self-identify as conservatives.92 Are we
seriously to assume, for example, that not only is there a
liberal and conservative genotype, but that this genotype
which, according to AFH, counts for more than environ-
ment in determining a person’s political orientation, is
somehow associated with income and with race? And if we
do not assume this, what possible genetic explanation could
account for such correlations?
But such correlations are well explained when we look
at the history of the United States, just as the rise and
transmission of liberalism and conservatism are well-
understood when we examine European history. In sum,
we can understand the world of political ideologies by
understanding history and culture, the ultimate “environ-
mental influences.” What, if anything, the study of genet-
ics adds to our understanding of political ideologies remains
to be seen.
Notes
1 Jonathan Beckwith, American Cancer Society Pro-
fessor of the Department of Microbiology and Mo-
lecular Genetics at Harvard University, 2006, 89.
Here is a small sampling (most of the reports listed
here have been published in multiple news sources
䡬 and news services generally supply stories to hun-
dred of newspapers worldwide): “‘Angry gene’
Could Help Spur Hostility,” Washington Post, March
9, 2007, reporting on a study by I. Halder, Univer-
sity of Pittsburgh; J. Grisolia, Scripps Mercy Hospi-
tal; and E. C. Suarez, Duke University; “Born to be
Bad? Genetic Research Says Maybe” Reuters, Febru-
ary 7, 2007, reporting on Harden et al. 2007; “Vio-
lence Is Blamed on ‘Warrior Gene’ in the Maoris,”
Daily Telegraph, August 10, 2006, reporting on
research of genetic epidemiologist Ron Lea (for the
controversy sparked by Lea’s claims see, e.g., “Scien-
tist Defends Warrior Gene,” New Zealand Herald,
March 5, 2007; “Maori Gene Claim Stirs Family
Violence Debate in New Zealand,” China Daily and
Global News Wire—Asia Africa Intelligence Wire,
August 15, 2006; “New Zealand’s Maori Minister
Hone Harawira Arguing that severe poverty and
high unemployment are to blame for the violence in
Maori communities, and not a “warrior” gene as
claimed by geneticist Rod Lea,” New Scientist, Au-
gust 19, 2006); “Lives of Crime,” Prospect, July 27,
2006, reporting on Caspi et al. 2002, “Role of
Genotype in the Cycle of Violence in Maltreated
Children”; “Conservative or liberal? It may be in the
genes,” November 2, 2006, Associated Press, report-
ing on Alford, Funk, and Hibbing 2005; “Some
Politics May Be Etched in the Genes,” New York
Times, June 21, 2005, reporting on Alford, Funk,
312 Perspectives on Politics
䡬
and Hibbing 2005; “Genes Contribute to Religious
Inclination,” Popular Scientist, March 16, 2005,
reporting on Koenig et al. 2005; “Study Finds Ge-
netic Basis For Human Kindness,” ABC News, Janu-
ary 12, 2005, reporting on Rushton 2004; “Israeli
Researchers Find ‘Altruism Gene,’” Jerusalem Post,
January 20, 2005, reporting on Bachner-Mehlman
et al. 2005; “Homosexuality is Genetic,” CanWest
News Agency, January 29, 2005, reporting on Mus-
tanski et al. 2005; “God Gene Discovered by Scien-
tist,” Telegraph Group Limited, reporting on Hamer
2004; “Of Genes and Exit Polls: Overlooked in
recent election commentary was the real reason for
the red-state victory. Heavy hint: It’s biological,”
Dan Seligman, Forbes, December 13, 2004, report-
ing on Rice and Hibbing 2004, Jost et al. 2003, and
Bouchard et al. 2003; “Happiness is Mostly Ge-
netic,” Forbes, September 23, 2004, reporting on
Lykken 2000; “Smoking Linked to Anger Gene,”
Herald Sun, February 14, 2004, reporting on Fallon
et al. 2004; “British Scientists Discover Criminal
Gene,” ABC News, August 5, 2002. In point of fact,
not a single gene has been discovered for any of
these traits, attitudes, or beliefs.
2 Galton 1875, 397.
3 Hamer and Copeland 1998, 22.
4 For altruism, see, e.g., Rushton, Littlefield, and
Lumsden 1986; for criminality, see, e.g., Moffitt
2005, Rhee and Waldman 2002; for “religiosity” see,
e.g., Bouchard et al. 1999; for conservatism, see,
e.g., Bouchard et al. 2003.
5 Alford, Funk, and Hibbing 2005.
6 Ibid., 164.
7 Ibid., 164.
8 Sigelman 2006, 172. Some of the excitement this
article has generated amongst political scientists is
doubtless due to the hope that at long last, the
dream of some political scientists to have political
science become a “true” science on the model of the
natural sciences—or rather, on the basis of a some-
what crude conception as to what makes the natural
sciences “science”—is about to be realized.
9 See the citations given in n. 1.
10 For the heritability of “criminal behavior” see,
e.g. Harden et al. 2007; Moffitt 2005; Caspi et al.
2002.
11 Elkins 2003, 296. See, e.g., “The Impact of Behav-
ioral Genetics on the Criminal Law,” special issue of
Law and Contemporary Problems 69 Winter/Spring
(2006); Paul S. Appelbaum, “Law & Psychiatry:
Behavioral Genetics and the Punishment of Crime,”
56 Psychiatric Services 25, 25 (2005); M. Jones,
“Overcoming the Myth of Free Will in Criminal
Law: The True Impact of the Genetic Revolution,”
52 Duke Law Journal 1031, 1039–40 (2003).
䡬
| |
䡬
12 Herrnstein and Murray 1994. See, e.g., Jacoby and
Glauberman 1995; Daniels, Devlin, and Roeder
1997; Gottfredson 1998, 1994.
13 Cooper and Zubek 1958.
14 For more on this kind of gene-environment inter-
action, see, e.g., McClearn 2004, Meaney 2001, Ram-
pon et al. 2000, Kempermann et al. 1997. The results
of this experiment contradict a central assumption
of A. R. Jensen 1969, set forth in a famous paper, “How
Much Can We Boost IQ and Scholastic Achieve-
ment?” Jensen claimed that IQ will in fact vary over envi-
ronments in the following manner: In impoverished
environments all genotypes will do equally poorly,
but in enriched environments the higher IQ genotypes/
phenotypes will always surpass the lower ones. For
a further critique of the assumptions of Jensen, see
Lewontin 2000, 22–30.
15 Shonkoff and Phillips 2000, 48.
16 There is speculation that the current “lag” in the
height of Americans is due to their “junk food” diet;
Komlos and Bauer 2003.
17 Plomin and Kosslyn 2001, 1153. See Meaney 2001,
50: “It is interesting to note that the most fervent prac-
titioners of quantitative human behavioral genetics
come not from genetics or even from biology at all.
They are very commonly persons trained in psychol-
ogy. It is unfortunate that psychology has exerted a much
greater effort in training its students in statistics
than in biology—with the predictable results. It is
really nothing more than a convenient myth to believe
that one can study the influence of genes without
understanding anything of molecular genetics.”
18 Gerald E. McClearn, Evan Pugh Professor of Health
and Human Development and Biobehavioral
Health, Department of Biobehavioral Health, Cen-
ter for Developmental and Health Genetics, Penn-
sylvania State University, 2004, 124.
19 Michael J. Meaney, James McGill Professor, Depart-
ments of Psychiatry and Neurology & Neurosurgery,
Program for the Study of Behavior, Genes and Envi-
ronment, Douglas Hospital Research Centre, McGill
University, 2001, 51.
20 Douglas Wahlstein, Great Lakes Institute for Envi-
ronmental Research and Professor of Biological
Sciences, University of Windsor, former President of
the International Behavioural and Neural Genetics
Society (was presented with the Distinguished Scien-
tist Award of the International Behavioural and
Neural Genetics Society in 2006 for “outstanding
lifetime achievement and contributions to the field
of behavioural and neural genetics”); Wahlstein and
Gottleib 1997, 178.
21 Johnson 2001.
22 Bruer and Greenough 2001; Nelson 2000.
23 Hurford 1991.
䡬
24 Karmiloff-Smith 2006, 48 (emphasis added in the
first sentence); see Wyman 2005, McClearn 2004,
Kahn 2003.
25 Parens, Chapman, and Press 2006, xvi.
26 Plomin 2004.
27 Hettema, Neale, and Kendler 1995, 327.
28 Evans and Martin 2000, 77; see Horowitz et al.
2003, Sandbank 1999, Ainslie 1997.
29 See, e.g., LaBuda et al. 1997, Kendler et al. 1994.
30 Joseph 2000, 544–45; 2004, 54–60, 69–70; cf.
Richardson and Norgate 2005, 34.
31 Scarr and Carter-Saltzman 1979, 528; Joseph 2000.
32 Eagly, Ashmore, Makhijani, and Longo 1991; Fein-
gold 1992; and Webster and Driskell 1983 in Perlini
et al. 2001, 277.
33 Burns and Farina 1992 and Buss 1994, in Horowitz
et al. 2003, 113–14.
34 Guo 2001, 122.
35 Bouchard 1993; Kendler et al. 1994; Rose 1991.
36 Joseph 2000, 545.
37 Bouchard and McGue 2003; Bouchard et al. 1990.
38 Studies have shown that parents’ recounts of their
rearing practices are often biased to match some
ideal of parenting; Bradburn, Rips, and Shevell
1987; Robbins 1963.
39 O’Connor and Deater-Deckard 1998; Ge and
Chadoret 1996; McCartney, Harris, and Bernieri 1990. 䡬
40 This is the claim made by Alford, Funk, and Hib-
bing 2005, 155.
41 Penderson et al. 1992; Bouchard et al. 1990; Lang-
invainio, Koskenvuo, and Kaprio 1984.
42 See, e.g., Bruer and Greenough 2001.
43 Joseph 2004, 131. Consider the following hypothet-
ical experiment: Of a group of 100 people, 50 are
told that it has just been discovered that they have
identical twins. These 50 meet their “identical
twins,” who are in fact members of the “study team”
recruited to be the same age and gender and look
almost identical to each of the 50 participants. For
one week they spend all of their time together to get
to know one another as well as possible. The 50
other participants are introduced to 50 other study
team members of the same age and gender to whom
they bear no particular physical resemblance, and are
told that they are strangers whom they should spend
one week trying to get to know well. Is it not highly
likely that the 50 “twins,” when tested, would self-
report significantly higher concordances to their fake
twins than the remaining 50 with their matched
“strangers”?
44 Beckwith 2006, 81; see Kamin and Goldberger
2002; Joseph 2004, 125–136.
45 Rowe 1994; Bouchard et al. 1990.
46 Beckwith 2006, 82; cf. Kamin and Goldberger
2002, 86.
June 2008 | Vol. 6/No. 2 313
| |
Exchange | Genes and Ideologies
47 See Alford, Funk, and Hibbing 2005, 155 n.3.
48 Kamin and Goldberger 2002, 87.
49 Beckwith 2006, 82.
50 Eckert, Heston, and Bouchard 1981, 180.
51 Joseph 2004, 98–104; Kamin and Goldberger 2002,
85; Watson 1982, 48.
52 The study of AFH (2005, 158–61) employs the
statistical method of polychoric analysis, whereas
most recent twin studies employ a more involved
statistical technique called Structural Equation Mod-
eling (SEM); Tomarken and Waller 2005. SEM,
however, in no way obviates all of the problems
associated with twin studies discussed above: It
depends every bit as much as polychoric analysis on
the Equal Environment Assumption, unbiased sam-
ples, and so forth. As Hettema, Neale, and Kendler
1995 (327) note: “Both traditional analysis of twin
data (e.g., Falconer 1989) and more recent struc-
tural equation models for twin analysis (Neale and
Cardon 1992) are predicated on the equal-
environment assumption (EEA)—that monozygotic
(MZ) and dizygotic (DZ) twins are equally corre-
lated for their exposure to environmental influences
that are of etiologic importance to the trait or disor-
der under study.” As Tomarken and Waller 2005
(56) note of the use of SEM in twin studies, “so-
䡬 phisticated statistical procedures cannot rescue a
poorly designed study.” As I have argued, twin stud-
ies are poorly designed studies; their flaws are of
such a nature that they cannot be rescued by statisti-
cal procedures applied to the data after the data has
been collected. I am indebted to an anonymous
reviewer for raising the issue of SEM, and to Ian
Morgan for an instructive exchange of e-mail on this
topic.
53 See e.g., Holsinger 1981, 144–5: “Since Occam’s
Razor ought to be invoked only when several hy-
potheses explain the same set of facts equally well, in
practice its domain will be very limited . . . Cases
where competing hypotheses explain a phenomenon
equally well are comparatively rare.”
54 I would like to emphasize, yet again, that this is
what my argument concerns: political ideology.
55 The study by Bouchard et al. 2003 was a study of a
small sample of twins raised apart—54 pairs of
monozygotic and 46 pairs of dizygotic twins reared
apart, whereas the study of AFH is a study of a large
sample of twins raised together.
56 AFH 2005, 158
57 Ibid., 164.
58 Ibid., 159.
59 Ibid., 159.
60 The importance of the constant refrain “at the
present time in the Unites States” should become
clear in what follows.
314 Perspectives on Politics
䡬
61 AFH 2005, 164.
62 Ibid., 158.
63 Ibid., 154, emphasis added.
64 Ibid., 163, emphasis added.
65 Ibid., 168.
66 Ibid., 153, 164.
67 Heywood 1992, 16–17.
68 Ryan 1993; Quinton 1993.
69 D’Souza 2002, 4–5.
70 E.g., Hobhouse 1964.
71 See, e.g., Kennedy 1999.
72 Waldron 1993, 36.
73 Ryan 1993, 291.
74 AFH 2005, 164–5.
75 In AFH 2005 they do not employ the expressions
“pro” and “con” attitudes, but I do not think they
would object to my use of them. The actual lan-
guage employed by AFH in describing the attitudes
that comprise the liberal-conservative phenotypes is
revealing. They scrupulously avoid the use of the
terms “moral,” “ethical,” “should,” “ought,” “right”
and “wrong;” but rather, in describing persons’
(moral and political) attitudes employ expressions
such as “desire for,” “fondness for,” “yearning for,”
and “preference for.” Thus, they write of a “fondness
for swift and severe punishments,” “a yearning for
in-group unity,” “a distaste for preset punishments,”
and “an aversion to inequality” (164–5). In char-
acterizing “easy” moral/political issues (164), in the
sense of issues about which people have very strong and
unambiguous opinions, the authors describe per-
sons’ strong beliefs as “gut responses.” Characterizing
moral/political beliefs by terms such as “fondness,”
“yearning,” “distaste,” “desire,” and “aversion,” makes
them appear equivalent to, e.g., a fondness for choc-
olate or an aversion to the smell of blue cheese. To write
of “gut responses” makes moral reactions seem on a
par with visceral reactions like nausea. Using the lan-
guage of “tastes” and “desires” to refer to complex
moral and political beliefs makes genetic explanations
of these beliefs seem less counter-intuitive. The lan-
guage of “tastes” and “desires” is well suited to genetic
explanations of human behavior, since it seems to
make more sense to talk of a genetically determined aver-
sion to, e.g., certain odors, then to talk of a geneti-
cally determined political ideology or moral belief
system. As is well known, the problem with the lan-
guage of tastes, desires, and aversions is that it fails to
account for two central features of morality as expe-
rienced by moral agents: First, that for the sake of doing
what one believes to be the morally right thing one
must at times act against his or her strongest “yearn-
ings,” “desires,” and “aversions”; and second, that rea-
son plays a role in moral judgment. The denial of any
role for reason in moral judgment is implicit in most
䡬
| |
䡬
of what AFH say about political attitudes and can be
seen quite clearly in their characterization of how
one’s moral-political beliefs might change (although
never using the term “moral”): “An individual with
a contextualist genotype who has been repeatedly vic-
timized by an out-group, or who has simply spent a
great deal of time listening to persuasive absolutists, may
adopt attitudes that run against type” (165). They
also comment that political ideologies are more diffi-
cult to “manipulate” to the extent that they are inher-
ited (164).Thus, according to the authors, moral beliefs
can change as a result of repeated negative
reinforcement—as a mouse might avoid a certain
type of food after receiving repeated electrical shocks,
“persuasion”—as opposed to argumentation, and
“manipulation.”
76 Madison et al. 1989, 58–9; cf. Shklar 1984.
77 See, e.g., Smart 1973.
78 Burke 1999; Quinton 1993.
79 See, e.g., Lasch 1991.
80 AFH 2005, 164.
81 Oddly enough, at one point AFH (ibid., 165) ap-
pear to clearly acknowledge the importance of cul-
ture in explaining human behavior: “Gene–culture
interaction is the key to understanding the source of
political attitudes and behaviors, just as it is the key
to understanding most physical and behavioral
aspects of the human condition. Genes do not work
in isolation and instead generally influence the ex-
tent to which organisms are responsive to particular
environmental conditions.” The problem is that
almost everything else they say in their article ig-
nores or contradicts this assertion.
82 Ibid., 154.
83 Ibid., 168.
84 Ibid., 154.
85 In fact, it is not clear that there could be such a
thing as “meaning” at all on a view in which discrete
or atomic attitude units cohered for “meaningless”
reasons. Nor is the idea of a discrete attitude unit
coherent in and of itself, since ideas (or “attitudes”)
can only be defined (or have meaning) in relation to
other ideas. To talk of a discrete or atomic attitude
unit is like proposing that a single sentence in a
given language could be comprehensible (or have
meaning) in isolation from the language system of
which it is a part.
86 Searle 1995.
87 Rutter 1996, 266.
88 AFH 2005, 157.
89 Ibid., 165.
90 Ibid., 157.
91 It is beyond the scope of this paper to consider the
relationship, if any, between personality and political
ideology. I shall simply note the following, which is
䡬
relevant to the present discussion. There exists what
might be called (for want of a better term) a “liberal
bias” in assorted assessments of personality, a bias
that is particularly evident in the Five-Factor Model
(FFM) and its characterization of the trait of “open-
ness.” FFM is based on the idea that the following
five dimensions are necessary and sufficient for
characterizing human personality: introversion/
extroversion; agreeableness; openness to experience;
conscientiousness; and emotionality/neuroticism.
The trait of openness is itself broken down into
sub-categories used to measure or “score” overall
“openness,” and one of these subcategories is “liber-
alism” (as measured by specifically political indica-
tors of liberalism); Costa and McCrae 1992. How
one scores on the “liberalism” category can signifi-
cantly affect how one scores on overall level of
“openness.” Thus, in a wholly circular fashion,
openness is correlated with liberalism by definition.
The concomitant of such a “liberal bias” in personal-
ity assessment is what might be called the “patholo-
gizing” of conservatism; Bailey 2004. For example,
in an article entitled “Political Conservatism as
Motivated Social Cognition,” the authors conclude
on the basis of a meta-analysis of a large number of
personality test results that “several psychological
variables predict conservatism.” It is positively corre- 䡬
lated with death anxiety, system instability,
dogmatism-intolerance of ambiguity, fear of threat
and loss; and negatively correlated with integrative
complexity, uncertainty tolerance, self-esteem and,
not surprisingly, openness to experience; Jost et al.
2003, 339. Research psychologist Robert Altemeyer,
inventor of the “Right Wing Authoritarian Scale”
(RWA), flatly denies that “an authoritarian impres-
sively like the authoritarian on the right reposes on
the left end of the RWA scale.” Those who score low
on the RWA scale are described as “fair-minded,
even-handed, tolerant, nonaggressive persons . . .
They are not self-righteous; they do not feel superior
to persons with opposing opinions”; Altemeyer
1981. To be sure dogmatic, intolerant, arrogant
“personality types” must be conservative and cannot
be liberal if such a stipulation is included in the
definition of “authoritarian,” “dogmatic,” and “ag-
gressive.” But it is worth asking whether we are
striving to give words a denotation that reflects our
own political ideologies, or seeking to uncover a
meaningful connection between personality traits
and the holding of particular political ideologies? If
we can successfully expunge all the prejudicial as-
sumptions that link particular personality traits with
particular political ideologies, there will be nothing
at all problematic about, e.g., a dogmatic, arrogant,
and self-righteous liberal.
June 2008 | Vol. 6/No. 2 315
| |
Exchange | Genes and Ideologies
92 Figures from the ANES Guide to Public Opinion
and Electoral Behavior, 1972–2004.
References
Ainslie, R.C. 1997. The Psychology of Twinship. North-
vale, NJ: Jason Aronson.
Alford, J.R., C. Funk, and J. Hibbing. 2005. Are politi-
cal orientations genetically transmitted? American
Political Science Review 99 (2): 153–67.
Alford, J.R., and J.R. Hibbing. 2004. The origin of
politics: An evolutionary theory of political behavior.
Perspectives on Politics 2 (4): 707–23.
Altemeyer, R.A. 1981. Right-Wing Authoritarianism.
Winnipeg: University of Manitoba Press.
Bachner-Mehlman, R.I. Gritsenko, L. Nemanov, A.H.
Zohar, C. Dina, and R.P. Epstein. 2005. Dopaminer-
gic polymorphisms associated with self-report mea-
sures of human altruism: A fresh phenotype for the
dopamine D4 receptor. Molecular Psychiatry 10:
333–35.
Bailey, R. 2004. “Pathologizing Conservatism.” Reason
Online (http://www.reason.com/rb/rb/102004.shtml),
October 20.
Beckwith, Jonathan. 2006. Whither human behavioral
genetics? In Wrestling with Behavioral Genetics: Sci-
䡬 ence, Ethics, and Public Conservation, ed. E. Parens, A.
Chapman, and N. Press. Baltimore: Johns Hopkins
University Press.
Bouchard, T. 1993. Genetic and environmental influ-
ences on adult personality: Evaluating the evidence.
In Basic Issues in Personality, ed. J. Hettema and I.
Deary. Dordrecht, The Netherlands: Kluwer.
Bouchard, T.J., Jr., D.T. Lykken, M. McGue, N.L.
Segal, and A. Tellegen. 1990. Sources of human psy-
chological differences: The Minnesota study of twins
reared apart. Science 12 (October): 223–28.
Bouchard, T.J,. and M. McGue. 2003. Genetic and
environmental influences on human psychological
differences. Journal of Neurobiology 54: 4–45.
Bouchard, T.J., Jr., M. McGue, D.T. Lykken, and A.
Tellegen. 1999. Intrinsic and extrinsic religiousness.
Twin Research 2 ( June): 88–98.
Bouchard, T.J., Jr., N.L. Segal, A. Tellegen, M. McGue,
M. Keyes, and R. Krueger. 2003. Evidence for the
construct validity and heritability of the Wilson-
Patterson Conservatism Scale: A reared-apart twins
study of social attitudes. Personality and Individual
Differences 34: 959–69.
Bradburn, N.M., L. Rips, and S.K. Shevell. 1987.
Answering autobiographical questions: The impact
of memory and inferences on surveys. Science 236:
157–61.
Bruer, J.T., and W.T. Greenough. 2001. The subtle
science of how experience affects the brain. In Critical
316 Perspectives on Politics
䡬
Thinking About Critical Periods, ed. B. Donald, J.
Bailey, Jr., J.T. Bruer, J.W. Lichtman, and F.J. Sy-
mons. Baltimore: Brookes Publishing Co.
Burke, Edmund. 1999. Reflections on the Revolution in
France. Select Works of Edmund Burke, vol. 2. India-
napolis: Liberty Fund.
Burns, G.L., and A. Farina. 1992. The role of physical
attractiveness in adjustment. Genetic, Social, and
General Psychology Monographs 118: 157–94.
Buss, D.M. 1994. The Evolution of Desire: Strategies of
Human Mating. New York: Basic Books.
Caspi, A., J. McClay, T. Moffitt, J. Mill, J. Martin, I.
Craig, A. Taylor, and R. Poulton. 2002. Role of geno-
type in the cycle of violence in maltreated children.
Science 297 (5582): 851–54.
Cooper, R.M. and J.P. Zubek. 1958. Effects of enriched
and restricted early environments on the learning
ability of bright and dull rats. Canadian Journal of
Psychology 12: 159–164.
Costa, P.T., Jr., and R.R. McCrae. 1992. NEO PI-R:
Professional Manual. Odessa, FL: Psychological Assess-
ment Resources.
Daniels, M., B. Devlin, and R. Roeder. 1997. Of genes
and IQ. In Intelligence, Genes, and Success: Scientists
Respond to the Bell Curve, ed. B. Devlin B., S.E. Fien-
berg, D.P. Resnick, and K. Roeder. New York:
Copernicus.
D’Souza, D. 2002. Letters to a Young Conservative. New
York: Basic Books.
Eagly, A.H., R.D. Ashmore, G.M. Makhijani, and
L.C. Longo. 1991. What is beautiful is good, but. . . :
A meta-analysis of research on the physical attrac-
tiveness stereotype. Psychological Bulletin 110:
109–128.
Eckert, E.D., L. Heston, and T.J. Bouchard. 1981. MZ
twins reared apart: Preliminary findings of psychiatric
disturbances and traits. Twin Research 3, Part B, ed.
L. Gedda, P. Parisi, and W.E. Nance. New York: A. R.
Liss.
Elkins, Lindsay. 2003. Five foot two with eyes of blue:
Physical profiling and the prospect of a genetics-based
criminal justice system. Notre Dame Journal of Law
Ethics & Public Policy 17: 269.
Evans, D.M., and N.G. Martin. 2000. The validity of
twin studies. GeneScreen 1: 77–79.
Falconer, D.S. 1989. Introduction to Quantitative Genet-
ics, 3d ed. New York: John Wiley and Sons.
Fallon, J.H., D. Keator, J. Mbogori, and S.G. Potkin.
2004. Hostility differentiates the brain metabolic
effects of nicotine. Cognitive Brain Research 18:
142–148.
Feingold, A. 1992. Good-looking people are not what
we think. Psychological Bulletin 111: 304–41.
Galton, W. 1875. The history of twins as a criterion of
the relative powers of nature and nurture. Journal of
䡬
| |
䡬
the Anthropological Institute of Great Britain and Ire-
land 5: 391–406.
Ge, X., and R.J. Chadoret. 1996. The developmental
interface between nurture and nature: A mutual influ-
ence model. Developmental Psychology 32: 524–89.
Gottfredson, L.S. 1994. Egalitarian fiction and collective
fraud. Society 31 (3): 553–60.
_. 1998. “The General Intelligence Factor: Despite
some popular assertions, a single factor for intelli-
gence, called g, can be measured with IQ tests and
does predict success in life.” Scientific American (Win-
ter), 24–29.
Guo, Sun-Wei. 2001. Does higher concordance in
monozygotic twins than in dizygotic twins suggest a
genetic component? Human Heredity 51: 121–32.
Hamer, D. 2004. The God Gene: How Faith Is Hard-
wired into Our Brains. New York: Doubleday.
Hamer, D., and P. Copeland. 1998. Living with Our
Genes. New York: Anchor Doubleday.
Harden, K.P., E. Turkheimer, R.E. Emery, B.M.
D’Onofrio, W.S. Slutske, A.C. Heath, and N.G.
Martin. 2007. Marital conflict and conduct problems
in children of twins. Child Development 78 (1): 1–18.
Herrnstein, R.J., and C. Murray. 1994. The Bell Curve:
Intelligence and Class Structure in American Life. New
York: Free Press.
Hettema, J.M., M.C. Neale, and K.S. Kendler. 1995.
Physical similarity and the equal environment as-
sumption in twin studies of psychiatric disorders.
Behavior Genetics 25: 327–35.
Heywood, A. 1992. Political Ideologies. New York: St.
Martin’s Press.
Hobhouse, L.T. 1964. Liberalism. New York: Oxford
University Press.
Holsinger, K. 1981. Comment: The blunting of Oc-
cam’s Razor. Canadian Journal of Zoology 59 144–6.
Horowitz, A.V., T.M. Videon, M.F. Schwartz, and D.
Davis. 2003. Rethinking twins and environments:
Possible social sources for assumed genetic influences
in twin research. Journal of Health and Social Behavior
44 ( June): 11–129.
Hurford, J. 1991. The evolution of the critical period
for language acquisition. Cognition 40: 159–201.
Jacoby, R., and N. Glauberman. 1995. The Bell Curve
Debate. New York: Times Books.
Jensen, A.R. 1969. How much can we boost IQ and
scholastic achievement? Harvard Educational Review
39: 1–123.
Johnson, M.H. 2001. Functional brain development in
humans. Nature Reviews Neuroscience 2: 475–83.
Joseph, J. 2000. Not in their genes: A critical review of
the genetics of attention-deficit disorder. Developmen-
tal Review 20: 539–67.
_. 2004. The Gene Illusion. New York: Algora
Publishing.
䡬
Jost, J.T., J. Glaser, A.W. Kruglanski, F.J. Sulloway.
2003. Political conservatism as motivated social cog-
nition. Psychological Bulletin 129 (3): 339–75.
Kahn, D.M. 2003. Nature versus nurture revisited: An
old idea with a new twist. Progress in Neurobiology 70
(1): 33–52.
Kamin, L.J., and A.S. Goldberger. 2002. Twin studies in
behavioral research: A skeptical view. Theoretical
Population Biology 61: 83–95.
Karmiloff-Smith, Annette. 2006. Ontogeny, genetics,
and evolution: A perspective from developmental
cognitive neuroscience. Biological Theory 1 (1):
44–51.
Kempermann, G., H.G. Kuhn, and F.H. Gage. 1997.
More hippocampal neurons in adult mice living in an
enriched environment. Nature 386: 493–95.
Kendler, K.S., M.C. Neale, R.C. Kessler, A.C. Heath,
and L.J. Eaves. 1994. Parental treatment and the
equal environmental assumption in twin studies of
psychiatric illness. Psychiatric Medicine 24: 579–90.
Kennedy, D. 1999. Freedom from Fear: The American
People in Depression and War, 1929–1945. New York:
Oxford University Press.
Koenig, L.B., M. McGue, R.F. Krueger, and T.J.
Bouchard. 2005. Genetic and environmental influ-
ences on religiousness: Findings for retrospective
and current religiousness ratings. Journal of Personality 䡬
73 (2): 471–88.
Komlos, J., and M. Bauer. 2003. “From the Tallest to
(One of ) the Fattest: The Enigmatic Fate of the
American Population in the 20th Century.” CESifo
Working Paper Series No. 1028 (http://papers.
ssrn.com/sol3/papers.cfm?abstract_id⫽444501).
LaBuda, M.C., D.S. Svikis, and R. Pickens. 1997. Twin
closeness and co-twin risk for substance use disorders:
Assessing the impact of the equal environment as-
sumption. Psychiatry Research 70: 155–64.
Langinvainio, H., M. Koskenvuo, and J. Kaprio.
1984. Finnish twins reared apart. II: Validation of
zygosity, environmental dissimilarity, and weight and
height. Acta Geneticae Medicae et Gemellologiae 33:
251–58.
Lasch, C. 1991. The True and Only Heaven: Progress and
Its Critics. New York and London: Norton.
Lewontin, R. 2000. The Triple Helix: Gene, Organism,
and Environment. Cambridge, MA: Harvard Univer-
sity Press.
Lykken, D. 2000. Happiness: The Nature and Nurture of
Joy and Contentment. New York: St. Martin’s Press.
Madison, James, Alexander Hamilton, and John Jay.
1989. The Federalist, ed. Jacob E. Cooke. Middle-
town, CT: Wesleyan University Press.
McCartney, K., M.J. Harris, and F. Bernieri. 1990.
Growing up and growing apart. Psychological Bulletin
107: 226–37.
June 2008 | Vol. 6/No. 2 317
| |
Exchange | Genes and Ideologies
McClearn, G.E. 2004. Nature and nurture: Interaction
and coaction. American Journal of Medical Genetics
Part B: Neuropsychiatric Genetics 124B (1): 124–30.
Meaney, Michael J. 2001. Nature, nurture, and the
disunity of knowledge. Annals of the New York Acad-
emy of Sciences. New York: New York Academy of
Sciences.
Moffitt, T.E. 2005. Genetic and environmental influ-
ences on antisocial behaviors: Evidence from
behavioral-genetic research. In Advances in Genetics
55, ed. Jeffrey C. Hall. Amsterdam: Elsevier Aca-
demic Press.
Mustanski, B.S., M.G. Dupree, C.M. Nievergelt, S.
Bocklandt, N.J. Schork, and D.H. Hamer. 2005. A
genomewide scan of male sexual orientation. Human
Genetics 116: 272–278.
Neale, M.C., and L.R. Cardon. 1992. Methodology for
Genetic Studies of Twins and Families. Dordrecht, The
Netherlands: Kluwer Academic.
Nelson, C.A. 2000. Neural plasticity and human devel-
opment: The role of early experience in sculpting
memory systems. Developmental Science 3: 115–130.
O’Connor, T.G., and K. Deater-Deckard. 1998.
Genotype-environment correlations in late childhood
early adolescence. Developmental Psychology 34:
970–82.
䡬 Parens, E., A. Chapman, and N. Press, eds. 2006. Wres-
tling with Behavioral Genetics: Science, Ethics, and
Public Conservation. Baltimore: Johns Hopkins Uni-
versity Press.
Penderson, N.L., R. Plomin, J.R. Nesselraode, G.E.
McClearn. 1992. A quantitative genetic analysis of
cognitive abilities during the second half of the life
span. Psychological Science 3: 346–53.
Perlini, A.H., A. Marcello, S.D. Hansen, W. Pudney.
2001. Effects of male age and physical appearance on
evaluations of attractiveness, social desirability and
resourcefulness. Social Behavior and Personality 29 (3):
277–87.
Plomin, R. 2004. Nature and Nurture: An Introduction
to Human Behavioral Genetics. Belmont, CA:
Wadsworth/Thomson.
Plomin, R., and S.M. Kosslyn. 2001. Genes, brain
and cognition. Nature Reviews Neuroscience 4:
1153–54.
Quinton, A. 1993. Conservatism. In A Companion to
Contemporary Political Philosophy, ed. R. Goodin and
P. Petit. Oxford: Blackwell Publishers.
Rampon, C., C.H. Jiang, H. Dong, Y.P. Tang, D.J.
Lockhart, P.G. Schultz, J.Z. Tsien, and Y. Hu. 2000.
Effects of environmental enrichment on gene expres-
sion in the brain. Proceedings of the National Academy
of Sciences USA 97: 12880–84.
Rhee, S.H., and I.D. Waldman. 2002. Genetic and
environmental influences on antisocial behavior: A
318 Perspectives on Politics
䡬
meta-analysis of twin and adoption studies. Psycholog-
ical Bulletin 128 (May): 490–529.
Richardson, K., and S. Norgate. 2005. The equal envi-
ronment assumption of classical twin studies may not
hold. British Journal of Educational Psychology 75:
339–50.
Robbins, L.C. 1963. The accuracy of parental recall of
aspects of child development and of child rearing
practices. Journal of Abnormal and Social Psychology
66: 261–70.
Rose, R. 1991. Twin studies and psychosocial epidemi-
ology. In Genetic Issues in Psychosocial Epidemiology,
ed. M. Tsuang, K. Kendler, M. Lyons. New Bruns-
wick, NJ: Rutgers University Press.
Rowe, D.C. 1994. The Limits of Family Experience:
Genes, Experience and Behavior. New York:
Guilliford.
Rushton, J.P., C.H. Littlefield, and C.J. Lumsden.
1986. Gene-culture coevolution of complex social
behavior: Human altruism and mate choice. Proceed-
ings of the National Academy of Sciences 83 (19):
7340–43.
Rutter, Michael. 1996. Concluding remarks. In Genetics
of Criminal and Antisocial Behavior, ed. Gregory R.
Bock and Jamie A. Goode. New York: John Wiley
and Sons.
Ryan, A. 1993. Liberalism. In A Companion to Contem-
porary Political Philosophy, ed. R. Goodin and P. Petit.
Oxford: Blackwell Publishers.
Sandbank, A. 1999. Twin and Triplet Psychology. New
York: Routledge.
Scarr, S., and L. Carter-Saltzman. 1979. Twin method:
Defense of a critical assumption. Behavior Genetics 9:
527–42.
Searle, John R. 1995. The Social Construction of Reality.
New York, NY: The Free Press.
Shklar, J. 1984. The liberalism of fear. In Liberalism and
the Moral Life, ed. N. Rosenblum. Cambridge MA:
Harvard University.
Shonkoff, J.P., and D.A. Phillips, eds. 2000. From Neu-
rons to Neighborhoods: The Science of Early childhood
Development. Washington, DC: National Academy
Press.
Sigelman, L. 2006. Report of the editor of the American
Political Science Review. PS: Political Science and Poli-
tics 40 (1): 171–73.
Silventoinen, K., J. Kaprio, E. Lahelma, and M. Kosk-
envuo. 2000. Relative effect of genetic and environ-
mental factors on body height: Differences across
birth cohorts among Finnish men and women. Amer-
ican Journal of Public Health 90: 627–30.
Smart, J.C.C. 1973. An outline of a system of utilitarian
ethics. In Utilitarianism: For and Against, ed. J.C.C.
Smart and Bernard Williams. Cambridge: Cambridge
University Press.
䡬
| |
䡬
Smith, A. 1976 [1775]. The Wealth of Nations. Oxford:
Clarendon Press.
Tomarken, A., and N.G. Waller. 2005. Structural
equation modeling as a data-analytic framework for
clinical science: Strengths, limitations, and mis-
conceptions. Annual Review of Clinical Psychology 1:
31–65.
Wahlstein, D., and G. Gottleib. 1997. The invalid
separation of effects of nature and nurture: Lessons
from animal experimentation. In Intelligence, Heredity
and Environment, ed. R.J. Sternberg and E.
Grogorenko. New York: Cambridge University Press.
䡬
Waldron, Jeremy. 1993. Liberal Rights. Cambridge:
Cambridge University Press.
Watson, P. 1982. Twins: An Uncanny Relationship. New
York: Viking.
Weber, M. 1978. Economy and Society. Ed. G. Roth and
C. Wittich. Berkeley: University of California Press.
Webster, M., Jr., and J.E. Driskell, Jr. 1983. Beauty as
status. American Journal of Sociology 89: 140–65.
Wyman, R.J. 2005. Experimental analysis of nature-
nurture interactions. Journal of Experimental Zoology
and Comparative Experimental Biology 303 (6):
415–21.
June 2008 | Vol. 6/No. 2 319
| |
䡬

Exchange

Beyond Liberals and Conservatives to

Political Genotypes and Phenotypes
John R. Alford, Carolyn L. Funk, and John R. Hibbing

In the past, most political scientists have been oblivious to the growing empirical evidence challenging environmental determinism.
Professor Charney, apparently as a result of the fact that genes and the environment interact in a complex fashion, advocates that this
passive unawareness be replaced by active denial. Science, however, does not advance by avoiding important relationships merely
because they are complicated and, fortunately, science is not heeding Charney’s ideologically-based fears. Molecular geneticists,
often working in tandem with political scientists, are quickly moving beyond twin studies to identify the specific suites of genes and
biological systems that predict variation in core political preferences, whatever labels those preferences might be given in a particular
culture at a particular time. We sympathize with the fact that our empirical findings, like those of so many behavioral geneticists,
make Charney uncomfortable; still, his critique serves up nothing new—empirically or otherwise. Just as analyses of the roots of
sexual preferences cannot presumptively ignore genetics, neither can analyses of the roots of political preferences.

ind-body dualism allows that the physical body is and implications. Genetics in general and twin studies in

M the product of natural, presumably genetic forces

but insists that the source of the human mind is
mystical and fundamentally different. The mind is not
nature’s but our own creation, shaped as we navigate and
are influenced by historical and cultural realities that we
have constructed. The human mind, dualism asserts, unlike
all other aspects of life on our planet, transcends crude
biological processes. And what better example of a pris-
tine experiential, non-biological, uniquely-human phe-
nomenon than mass-scale politics with its relatively recent
advent and its constantly changing issues, terms, parties,
and players. For those caught in the hubris of this dualist
perspective, empirical evidence indicating that political
orientations are transmitted genetically as well as cultur-
ally seems patently “incoherent” and must be the product
of a flawed methodology.
Thus it is that Evan Charney rises to attack a recent
article of ours in the American Political Science Review, 1
an article he fears is part of “a trend among behavioral
scientists to view ever more complex attitudes as in some
sense genetically determined.” 2 We welcome the appear-
ance of Charney’s essay. The scientific process needs schol-
ars eager to question methodologies, data, interpretations,
John R. Alford is Associate Professor of Political Science,
Rice University (jra@rice.edu). Carolyn L. Funk is Associ-
ate Professor of Political Science, Virginia Commonwealth
University (clfunk@vcu.edu). John R. Hibbing is the Foun-
dation Regents Professor of Political Science, University of
Nebraska-Lincoln (jhibbing@unl.edu)
particular are not familiar topics to most political scien-
tists. Several of the misconceptions evident in Charney’s
written remarks are undoubtedly held by other members
䡬
of the discipline so we are pleased to have been offered the
opportunity to explain our procedures, findings, and con-
clusions more fully. Charney’s criticisms fall into two broad
categories. The first is that the methods we employed to
obtain our results—specifically, the classical twin design—
rest on faulty assumptions and therefore yield meaning-
less results. The second, drawing directly from a dualist
perspective, is that political beliefs are entirely embedded
in culture and therefore, logically could not have a genetic
component. We address each of these charges in turn.
Our central point is that it is insufficient merely to assert
the implausibility of political ideology being partially her-
itable; ultimately this matter can and must be decided by
the scientific process.
The Methodical Challenge
Much of Charney’s methodological critique centers on
what has come to be termed the “equal environments
assumption” (EEA). Monozygotic (MZ) twins share 100
percent of their genetic heritage while dizygotic (DZ) twins,
like all full siblings, share roughly 50 percent. This known
difference in genetic similarity across the two types of twins
provides an opportunity to estimate the importance of
genetic similarity—but only if the environments of MZ
and DZ twins are equally similar. If, relative to DZ twins,
MZ twins not only share more of their genetic code but
also share more of their environmental experiences, vari-
ance attributed to genetics may actually be the result of

doi:10.1017/S1537592708080638 June 2008 | Vol. 6/No. 2 321

䡬
| |
Exchange | Political Genotypes
environmental forces. Thus, the key concern is that heri-
tability estimates will be inflated while the effect of shared
environment will be underestimated.
The equal environments assumption is an important
concern and Charney is quite right to raise it, but the
EEA has been the subject of a tremendous amount of
debate and research and Charney does not provide a fair
assessment of current wisdom in the behavioral genetics
literature. We have space to offer only a few corrections.
One misconception springs from Charney’s treatment of
the environment as exogenous even as such a practice is
invalidated by his own argument. Charney invokes a com-
ment from Horwitz et al. pointing out that “because mono-
zygotic twins have greater physical resemblance to each
other than dizygotic twins, they could have a greater chance
of receiving similar social reactions.” 3 If, as Charney and
Horwitz believe, genes shape physical traits that in turn
shape social reactions that in turn shape beliefs, the link-
age structure is complete and genes will affect beliefs.
Whether the causal order is [genes r beliefs] or [genes r
physical traits r social reactions r beliefs], the underlying
cause is still genetic. An example of the endogeneity of
environmental forces is provided by O’Connor et al. who
report that adoptive children genetically at risk for anti-
social behavior are significantly more likely than not-at-
genetic risk adoptive children to be the recipient of negative
䡬 parenting from their adoptive parents.4 Negative parent-
ing is typically assumed to be the cause of children’s anti-
social behavior but in point of fact children play an
important role in shaping their own environment, in this
case by influencing the behavior of their parents.
Even if we pretend genes have nothing to do with shap-
ing the environment, the EEA withstands analysis. A sur-
prising number of the parents of twins, as many as 20–30
percent in some studies, mis-categorize the zygosity of their
twins, thus creating a valuable opportunity to distinguish
environmental from genetic influences. If Charney is
correct, the determining factor in twin pairs’ degree of sim-
ilarity should be their perceived zygosity; if we are correct,
the determining factor in degree of similarity should be their
actual zygosity.The results of these mis-categorization stud-
ies are clear: DZ twin pairs believed by their environments
to be MZ twin pairs are no more similar than DZ twin pairs
believed to be DZ twin pairs.5 This basic finding is impos-
sible to explain if the environment is all that matters.
MZ twins are indeed more likely to share certain envi-
ronmental experiences. They are, for example, more likely
than DZ twins to share the same bedroom and to have the
same friends but this fact in and of itself does not consti-
tute a fatal flaw for twin studies.6 The central question is
whether variable similarity on friends and bedrooms inde-
pendently influences the specific trait of interest—in this
case, political beliefs—and no evidence has yet been pre-
sented that it does. In fact, empirical evidence goes fur-
ther and demonstrates that the correlation between the
322 Perspectives on Politics
䡬
overall frequency of twin contact and the similarity of
their social and political attitudes is small and statistically
insignificant.7 In terms of politics this result is not all that
surprising since it seems unlikely that the parents of DZ
twins are more likely than the parents of MZ twins to
push each of their twins in a different political direction.
Compared to the parents of MZ twins, why would the
parents of DZ twins be less desirous that all of their off-
spring share the same political beliefs?
But rather than debate what is plausible or implausible
let us turn again to the empirical evidence. The most impor-
tant challenge we wish to issue to Charney concerning the
EEA involves the differential heritability estimates of polit-
ical attitudes on the one hand and party identification on
the other. Similar to previous research,8 we estimate the
heritability of political and social attitudes to be in the .4
to .5 range, leaving .5 to .6 attributable to environmental
factors. But these same procedures reveal that party iden-
tification is only about .14 heritable, leaving .86 attribut-
able to the environment, so the classical twin design reports
a dramatic difference in the heritability of political beliefs
and party identification. Interestingly, other research iden-
tifies precisely the same pattern for religious beliefs (strong
heritability) as opposed to religious denominational affil-
iation (minimal to no heritability).9 If violations of the
EEA are responsible for reported heritability, Charney must
argue that parents of MZ and DZ twins raise their chil-
dren equally similarly with regard to party identification
but differentially with regard to political attitudes. Why
would parents of DZ twins socialize their twins to have
the same party identification but different political beliefs,
while parents of MZ twins socialize their twins to have
both similar party identifications and similar political
beliefs?
Rather than merely relying on assertions, we have also
empirically estimated heritability while accounting for the
EEA. To do so, we have re-estimated our findings from
the Virginia 30K sample using more sophisticated struc-
tural equation models that account for shared environ-
mental influences from other family relationships—
namely parents and non-twin siblings.10 This re-estimation
with non-twin data allows us to correct for shared effects
among siblings and to estimate twin-specific environment
effects. The findings support our original claims that genet-
ics accounts for at least 40 percent of the variance in ideo-
logical orientations as measured by the 28-item Wilson-
Patterson index. A different approach is taken by Fowler,
Baker and Dawes in their tests of the heritability of voter
turnout.11 Fowler et al. test the comparability of MZ and
DZ pairs on a number of politically-relevant variables and
indicators of socioeconomic status. They find no mean
differences between MZ and DZ pairs on these political
and social variables, thus demonstrating that the trait-
specific environment is simply too equal to offer any power
in explaining away the heritability findings. Research in
䡬
| |
䡬
other areas has similarly shown either that the EEA is
valid or that violations of it are too modest to negate
results indicating the important influence of genes.12
Twin studies do have the potential to mislead. One of
Charney’s central points is that genes and the environ-
ment cannot be partitioned neatly (e.g., 14 percent genetic
and 86 percent environmental) since they interact in such
complicated ways. In so saying, however, Charney is only
parroting a point we stressed in our original article: “Gene
culture interaction is the key to understanding the source
of political attitudes and behaviors, just as it is the key to
understanding most physical and behavioral aspects of the
human condition. Genes do not work in isolation and
instead generally influence the extent to which organisms
are responsive to particular environmental conditions.” 13
Providing statistical estimates of the effects of genetics,
shared environment, and unshared environment is not
tantamount to asserting that these effects work in isola-
tion from each other. We know of no practitioner of twin
studies who believes they do. The interactions among genes
and between genes and the environment create an amaz-
ingly complex situation. The good news is that, since Char-
ney apparently believes genes and the environment interact
to influence phenotypes, he must agree with us that genes
are indeed behaviorally relevant.
Charney is also concerned with the fact that estimates
of heritability are specific to a given population in a given
environment and are therefore subject to substantial vari-
ation. This is an important point, and its import is often
misunderstood, as it is by Charney. Look carefully at his
example (one that can be reproduced in many other guises).
He describes a study that examines the heritability of maze
running ability in mice, utilizing mice selectively bred for
high or low ability on this trait. As Charney describes the
results:
For mice raised in a “standard” environment heritability of this
trait was high. The researchers then raised the offspring of the
genetically selected lines in two “extreme environments,” a cog-
nitively poor one (dull colors all around, no toys) and a cogni-
tively enriched one (bright colors and patterns, many toys). Both
the bright and dull lines behaved “stupidly” in the poor environ-
ment and “smartly” in the enriched one, with the result that the
heritability of the trait dropped to zero in both extreme environ-
ments. The bright and dull mice may well have inherited what-
ever genes are linked to intelligence (or maze running aptitude)
in mice (which is what they were bred for), but the heritability of
the trait of intelligence was shown to be highly dependent upon
post-natal environment.14
Several points are of critical importance here. First, notice
that Charney admits to the well-established fact that the
mice “may well have inherited whatever genes are linked
to intelligence.” Second, notice that when mice were raised
in a standard environment, estimates of heritability cor-
rectly detected the significant role of genes in influencing
maze-running ability. Only when mice were raised in arti-
ficially extreme environments did the apparent heritability
䡬
of intelligence drop to near zero. Actual genetic inheri-
tance continued to take place and the separate offspring of
the two distinct lines would still show divergent abilities if
raised in a standard environment. The extreme environ-
ments only inhibited the ability of a test of heritability to
detect this continued role of genetic transmission and the
direction of any bias in the estimate of heritability is con-
servative. In this study, extreme environments led to under-
estimation of the role genes play in the natural world, not
to over-estimation. Far from suggesting that we should
not put much faith in high estimates of heritably (such as
those we found for political orientations), this example
demonstrates the reverse. Studies finding zero heritability
in specific environments are the ones that may be suspect,
and can produce results that mask the true role of genes in
the transmission of behavioral traits.
In one important area of methodology we are in full
agreement with Charney. Twin studies should be seen as
the beginning and not the end of research in this area and
we are following this model in our own research. The twin
design is typically and most profitably employed as a pre-
cursor to additional genetic work. Wet genetic techniques
such as linkage analysis, which can isolate the location of
genes relevant to a phenotypic behavior of interest, are
best conducted with DNA drawn from close relatives such
as twins. Many twin registries now have data banks con-
taining the twins’ DNA because of the value in using the 䡬
same subjects to assess heritability levels and then to con-
duct genotyping work. Behavioral geneticists commonly
move seamlessly back and forth from the twin design to
bench genetics. We are currently working with several
geneticists including Nicholas Martin (Laboratory Head
of Genetic Epidemiology at the Queensland Institute of
Medical Research in Brisbane), Lindon Eaves (Professor
of Human Genetics at the Virginia Commonwealth Uni-
versity School of Medicine), and Shelley D. Smith (Direc-
tor of Molecular Genetics at the University of Nebraska
Medical Center) and have begun the process of statisti-
cally associating particular genetic alleles with political
beliefs and political behaviors, a technique called allelic
association.
Twin studies alone are no substitute for using linkage
analysis and allelic association studies to find the specific
genes contributing to the heritability of political attitudes,
but searching for specific genes would be a waste of time
absent evidence of heritability. With billions of nucleotide
base-pairs in the nucleus of most human cells, locating
the relevant individual genes is a challenging, multidisci-
plinary undertaking. Even when associations are found,
much work remains since correlations alone will not reveal
the complex pathway between genotype and phenotype.
No serious scholars doubt the difficulties attending efforts
to link genes to political behaviors and beliefs but we have
no doubts this link will eventually be established. Addi-
tional information on the role of evolution, emotions, the
June 2008 | Vol. 6/No. 2 323
| |
Exchange | Political Genotypes
limbic system, and neurotransmitters in shaping personal
and political traits emerges every week 15 and the pace is
only going to pick up, making it an uncomfortable time
for dualists such as Charney.
Charney writes that “Alford, Funk, and Hibbing have
not discovered a set of genes corresponding to their hypoth-
esis of genetic causation.” This is quite true, we have not.
But, as noted above, we are working on it. Scholars else-
where have uncovered genes involved with reading disor-
ders, depression, autism, risk-taking, and attention deficit
hyperactivity disorder. Expecting one article to identify
conclusively the specific genes and biological systems
responsible for the heritability of something as complex as
political beliefs is unfair. Achieving this goal will take
numerous scholars working for extended periods of time
in close inter-disciplinary cooperation and to say this effort
should not be undertaken because it is “complicated” is
both defeatist and anti-science.
Political Attitudes, Orientations,
and Ideologies
Charney’s critique of twin studies is predictable and largely
uninformed by the state of the art in behavioral genetics,
but the second half of his article provides the opportunity
for an important scholarly exchange on the nature of polit-
ical attitudes, political orientations, and political ideolo-
䡬 gies. Here Charney challenges our work on two grounds;
first a narrow issue concerning current U.S. ideological
divisions, and second a much broader conceptual cri-
tique. We will address the narrower issue first.
Charney contends that our findings from the twin study
are flawed because the collection of issues in the Wilson-
Patterson index holds no face validity as a measure of
political ideology and because they indicate the “wrong”
issues as being the most heritable. Specifically, the highest
heritability levels reported in our article are for property
taxes, school prayer, “Moral Majority,” capitalism, and
astrology. Charney believes, rightly, that this is a curious
collection of issues. He asserts that the highest heritability
should be for abortion, gay rights, and other hot-button
social issues related to sexuality and religion or that, at the
very least our index of conservatism should give more
weight to these issues.
Charney’s concerns here are largely a moot point, how-
ever. As mentioned in our article, assortative mating is
particularly powerful with regard to political and religious
beliefs (conservatives mate with conservatives and liberals
mate with liberals) and if parents are genetically similar
then simple twin designs such as the one we employed
will underestimate heritability.16 Applying the simplest pos-
sible correction for assortative mating identifies the fol-
lowing five issues, in order, as the most heritable:
1. Living together (.67)
2. School prayer (.66)
324 Perspectives on Politics
䡬
3. Abortion (.64)
4. X-rated movies (.63)
5. Gay rights (.60).
If Charney is correct that an acceptable test for estimating
heritability of political orientations must indicate high lev-
els of heritability for the social issues that he calls “the
central concerns dividing liberals and conservatives,” then
the twin study findings pass this test with flying colors.
On a broader front, Charney probably speaks for a
majority of political scientists in maintaining that politi-
cal attitudes are socially constructed, free-standing orien-
tations to political concepts and issues. Charney is especially
taken with the context-bound nature of words such as
liberal, conservative, left, and right. Since these terms are
culturally rooted, he seems to conclude that any orienta-
tion to politics must be culturally rooted. He spends much
of his essay recounting the history of these terms, showing
that what it means to be a liberal today in the United
States is not what it meant to be a liberal in France in the
seventeenth century. And, of course, for much of the world,
terms like “liberal” have no meaning at all. Given this,
Charney asks, how is it possible that there could be a
genetic basis for liberalism or any other political concept?
Our measure of political orientation consisted of an
additive index of the 28 Wilson-Patterson items available
to us from the Virginia 30K study (contrary to Charney’s
claims, we used all 28 items present in the survey and did
not pick some items and exclude others). We used these
items as an index scored to reflect a continuum from lib-
eral to conservative attitude positions. In choosing our
language for discussing political orientation we opted for
descriptors common to the field of political science, but
which are probably not the most apt descriptors for the
underlying phenotype. We believe that the actual under-
lying phenotype, whatever it is ultimately labeled, will be
applicable across time and place. As such, what we have
called “bedrock principles of group life” is more likely to
fit the bill. The empirical evidence showing a genetic com-
ponent for specific attitude positions, for example, is likely
to reflect an indirect role of genetics on attitudes through
the heritability of values or orientations more fundamen-
tal to living in social groups.
Recent work by Peter K. Hatemi and colleagues dem-
onstrates that vote choice is not directly heritable but rather
primarily indirectly heritable through political attitudes
toward issues such as abortion, school prayer, and the death
penalty.17 Up to now, detailed measures of social and polit-
ical orientations have not been administered to twins (a
situation we are in the process of remedying thanks to a
recent grant from the National Science Foundation). When
data on core preferences for social group life are available
we predict they will show that attitudes toward school
prayer, abortion, and the death penalty are themselves
indirectly heritable and that what is directly heritable are
䡬
| |
䡬
orientations to bedrock principles of group life. But, of
course, these ideas have yet to be tested and must be con-
sidered in comparison with alternative conceptualizations
of the underlying phenotype.
We have evidence in both the U.S. and Australia for a
sizeable genetic component in the transmission of politi-
cal orientations but this is not yet enough to claim with
confidence that these findings are applicable across all cul-
tures and contexts. Regardless, we do not agree with Char-
ney’s unsubstantiated assertion that culture universals do
not exist. Schwartz and Bardi’s recent cross-cultural study
of values more realistically captures the situation:
Studies at the national level reveal a great deal of variation in the
value priorities of individuals within societies as well as groups
across nations. The research suggests that individuals both within
and across societies have quite different value priorities that reflect
their different genetic heritage, personal experiences, social loca-
tions, and enculturation. Yet hidden behind these important dif-
ferences is a surprise that may reflect something about the origins
and role of values for human society. Researchers, including our-
selves, have focused almost exclusively on differences in value
priorities. When we switch our focus to ask about similarities,
we discover a striking degree of consensus across individuals and
societies . . . Differences are more salient and compelling than
similarities. It may therefore be difficult to accept that a largely
shared, pan-cultural value hierarchy lies hidden behind the strik-
ing value differences that draw our attention. Differences help us
to identify the influences of unique genetic heritage, personal
experience, social structure, and culture on value priorities. The
pan-cultural hierarchy points to the bases of values in shared
human nature and to the adaptive functions of values in main-
taining societies. To gain a full understanding of human value
priorities, we must take note of the interplay of both differences
and similarities.18
Charney rightly raises the critical question of just how
genes could influence political beliefs. Numerous possibili-
ties exist. The connection could run through personality;
it could run through temperaments such as risk-taking,
fear, and preference for structure and certainty; it could
run through a tendency to hold beliefs intensely or casu-
ally; or it could even run through genetically-based phys-
ical traits, as Charney speculates. There are endless
possibilities and we do not pretend to know the actual
mechanisms. This is what we are investigating now. As
mentioned above, our collaborative research with the goal
of associating particular genetic alleles with particular
political beliefs and behaviors will move us down that
path. But it will be a long and difficult process, requiring
collaborative efforts among scholars with an open mind
and a healthy respect for the scientific process of testing
and revising hypotheses based on empirical observation.
Led by numerous social scientists including Jost, Haidt,
and Thornhill, there is growing support for seeing political
ideology as springing from deeper, seemingly non-political
traits.19 Of course, researchers have long noted the connec-
tion between personality and politics and the evidence is
rapidly building that political ideology is intimately con-
䡬
nected with other aspects of temperament.20 For example,
compared to liberals, political conservatives have funda-
mentally different approaches to child-rearing.21 Our own
recent analysis finds that a simple four-item strict parent-
ing scale is strongly correlated with the Wilson-Patterson
index (r ⫽ .45; p ⬍ .001). Similarly, people who say “good
manners and cleanliness matter” are more likely to be polit-
ically conservative (gamma ⫽ .32; p ⬍ .001) and people who
“would take drugs that may have strange effects” are more
likely to be politically liberal (gamma ⫽ .48; p ⬍ .001).22
The brains of conservatives react differently, on average, from
the brains of liberals to an unexpected (but completely non-
political) stimulus.23 And it is possible to make accurate pre-
dictions of the political beliefs of adults on the basis of the
observed (not self-reported) social behaviors of those same
individuals at nursery school twenty years previously.24 In
sum, people vary dramatically in their orientation toward
order, threat, risk, uncertainty, and obedience to tradi-
tional values and these variations in turn are statistically
related to political attitudes and ideology. Ideology is con-
nected to foundational preferences for social and political
structure and these preferences likely have a basis in neuro-
transmitters, brain activity, and ultimately genetics.
We have hopes that raising the possibility that politics
is partially heritable will encourage a much needed recon-
sideration of the nature of political temperament. The
discipline would benefit from trying new research meth- 䡬
ods and from viewing political ideologies as something
other than completely “context dependent.” In our view,
if genes shape politics, the connection does not run
through the musings of nineteenth- and twentieth-
century political theorists but rather through the core
principals of social life as they have existed since before
we were fully human.
Conclusion
No scientifically literate person in this day and age can
claim that genes are irrelevant to human behavior and
predispositions, yet many people are deeply discomfited
by this reality. One common solution is to concede that
genes affect behavior, but to assert that the connection is
extraordinarily complex and then to proceed as if this
complexity negated any requirement that we incorporate
this sometimes uncomfortable reality into research into,
and understanding of, human behavior. This is where we
part company. To us, the reality of a complex link between
genes and political phenotypes means political scientists
must rein in their reflexive environmentalism and join
forces with behavioral geneticists and others in the biolog-
ical community to continue the exacting, time-consuming,
but deeply exciting genetic research that has already begun
the process of teasing out this complexity.
We never claimed that a focus that incorporates genes
obviates the need to also consider the environment. Indeed,
June 2008 | Vol. 6/No. 2 325
| |
Exchange | Political Genotypes
casting the issue as genes competing with the environ-
ment, as Charney does in his conclusion (“if genes count
for more than environment the phenomena of liberalism
and conservatism . . . become utterly incomprehensible”)
is silly and misses the point. What we claim is that genes
are important to political thought and behavior. We real-
ize that for a discipline as completely vested in nurture
alone as political science, the need to look at both nurture
and nature will constitute an important shift. As Charney
puts it:
The claim of Alford, Funk, and Hibbing is indeed astonishing,
because if true, it would require nothing less than a revision of
our understanding of all of human history, much, if not most of
political science, sociology, anthropology, and psychology, as well
as, perhaps, our understanding of what it means to be human.25
The “claim” is of course not wholly ours, but more
appropriately the claim of over twenty years of behavioral
geneticists who pioneered research in this area. But we are
in complete agreement with what is at stake. Our 2005
article found a substantial degree of heritability for polit-
ical orientations, but political behavior subtends far more
than political orientations and the study of genetic influ-
ences does not end with a single classical twin study. Genet-
icists can only go so far in their investigation of a field as
foreign to them as political behavior. Political scientists
need to engage if genetic research in this area is to progress
䡬 York: Harper.
beyond rudimentary dependent variables and if political
science is going to progress beyond a narrowly limited set
of environmental independent variables. Colleagues from
across the social and natural sciences are currently con-
ducting research on large-scale human social cooperation
that is among the most exciting and challenging scientific
revolutions of our times. With political scientists largely
isolated from this effort, it is no surprise that researchers
from other fields have mostly missed the point that large-
scale human cooperation is politics, not just neuroeconom-
ics or evolutionary psychology. We can right this wrong if,
and only if, we join with our colleagues from across the
sciences in this grand search.
Notes
1 Alford, Funk, and Hibbing 2005.
2 Charney 2008.
3 Horwitz et al. 2003, 113–14.
4 O’Connor et al. 1998.
5 Scarr and Carter-Saltzman 1979; Plomin 1990;
Kendler et al. 1993; Hatemi 2007.
6 Scarr and Carter-Saltzman 1979; Kendler et al.
1992.
7 Martin et al. 1986.
8 Ibid.
9 Bouchard and McGue 2003.
10 Hatemi et al. 2007b.
11 Fowler, Baker, and Dawes 2006.
326 Perspectives on Politics
䡬
12 E.g., Plomin et al. 1976; Morris-Yates et al. 1990;
Kendler et al., 1993; Xian et al. 2000; Cronk et al.
2002; Eriksson et al. 2006.
13 Alford, Funk, and Hibbing 2005: 165.
14 Charney 2008, 300-1.
15 Madsen 1986; Marcus, Neuman, and MacKuen
2000; Orbell et al. 2004; McDermott 2004; Schreiber
2005; Lodge and Taber 2005; Westen et al. 2006;
Alford and Hibbing 2007; Carmen 2007.
16 Hatemi et al. 2007b.
17 Hatemi et al. 2007a.
18 Schwartz and Bardi 2001.
19 Jost 2006; Jost et al. 2003; Haidt and Graham
2008; Thornhill and Fischer 2007.
20 Adorno et al. 1950; McClosky 1958.
21 For an early example see Laswell 1930, or more
recently Lakoff 2002.
22 Alford and Hibbing 2007.
23 Amodio, Jost, Master, and Yee 2007.
24 Block and Block 2005.
25 Charney 2008, 300.
References
Adorno, T., E. Frenkel-Brunswik, D. Levinson, and R.
Sanford. 1950. The Authoritarian Personality. New
Alford, John R., Carolyn L. Funk, and John R. Hib-
bing. 2005. Are political orientations genetically
transmitted? American Political Science Review 99 (2):
153–68.
Alford, John R., and John R. Hibbing. 2007. Personal,
inter-personal, and political temperaments. Annals of
the Academy of Political and Social Science 614 (No-
vember): 196–212.
Amodio, David M., John T. Jost, Sarah L. Master, and
Cindy M. Yee. 2007. Neurocognitive correlates of
liberalism and conservatism. Nature Neuroscience 10:
1246–7.
Block, Jack, and Jeanne H. Block. 2005. Nursery school
personality and political orientation two decades later.
Journal of Research in Personality 39: 395–422.
Bouchard, T.J., Jr., and Matt McGue. 2003. Genetic
and environmental influences on human psychologi-
cal differences. Journal of Neurobiology 54: 4–45.
Carey, Gregory. 2003. Human Genetics for the Social
Sciences. Los Angeles, CA: Sage Publications.
Carmen, Ira H. 2007. “From Genes to Mind to
Politics.” Presented at the annual meeting of the
Midwest Political Science Association, Chicago,
April 12–15.
Charney, Evan. 2008. Genes and ideologies. Perspectives
on Politics 6 (2): 299–315.
Cronk, N.J., W.S. Slutske, P.A. Madden, K.K Bucholz,
W Reich, and A.C. Heath. 2002. Emotional and
䡬
| |
䡬
behavioral problems among female twins: An evalua-
tion of the equal environments assumption. Journal of
the American Academy of Child and Adolescent Psychia-
try 41: 829–37.
Eriksson, Marit, Finn Rasmussen, and Per Tynelius.
2006. Genetic factors in physical activity and the
equal environment assumption: The Swedish young
male twins study. Behavior Genetics 36: 238–47.
Fowler, James H., Laura Baker, and Christopher T.
Dawes. 2006. “The Genetic Basis of Political Cooper-
ation.” Presented at the annual meeting of the Ameri-
can Political Science Association, Philadelphia, PA,
August 30–September 2.
Gerring, John 1997. Ideology: A definitional analysis.
Political Research Quarterly 50: 957–94.
Gosling, Samuel D., and Oliver P. John. 1999. Personal-
ity dimensions in non-human animals: A cross-species
review. Current Directions in Psychological Science 8:
69–75.
Haidt, Jonathan, and Jesse Graham. 2008. “When
Morality Opposes Justice: Conservatives Have Moral
Intuitions that Liberals May Not Recognize.” Social
Justice Research 21, forthcoming.
Hatemi, Peter K. 2007. “The Genetics of Political Atti-
tudes.” PhD Diss., University of Nebraska-Lincoln.
Hatemi, Peter K., Sarah E. Medland, Kate I. Morley,
Anthony C. Heath, and Nicholas G. Martin. 2007a.
The genetics of voting: An Australian twin study.
Behavior Genetics 37: 435–48.
Hatemi, Peter K., John R. Alford, John R. Hibbing,
Nicholas G. Martin, and Lindon J. Eaves. 2007b.
“Not by Twins Alone: Using the Extended Twin
Family Design to Investigate the Genetic Basis of
Political Beliefs.” Presented at the annual meeting of
the Midwest Political Science Association, Chicago,
April 12–15.
Horwitz, Allan, Tami Videon, Mark Schmitz, and Diane
Davis. 2003. Rethinking twins and environments:
Possible social sources for assumed genetic influences
in twin research. Journal of Health and Social Behavior
44: 111–29.
Jost, John T. 2006. The end of the end of ideology.
American Psychologist 61: 651–70.
Jost, John T., Jack Glaser, Arie W. Kruglanski, and
Frank J. Sulloway. 2003. Political conservatism as
motivated social cognition. Psychological Bulletin 129:
339–75.
Kendler, Kenneth S., Michael C. Neale, R.C. Kessler,
Andrew C. Heath, and L.J. Eaves. 1992. A popula-
tion based twin study of major depression in women:
The impact of varying definitions of illness. Archives
of General Psychiatry 49: 257–66.
_. 1993. A test of the equal environment assump-
tion in twin studies of psychiatric illness. Behavior
Genetics 23: 21–27.
䡬
Lakoff, George. 2002. Moral Politics: How Liberals and
Conservatives Think. Chicago: University of Chicago
Press.
Laswell, Harold. 1930. Psychopathology and Politics.
Chicago: University of Chicago Press.
Lodge, Milton, and Charles S. Taber. 2005. The auto-
maticity of affect for political leaders, groups, and
issues. Political Psychology 26: 455–82.
Madsen, Douglas. 1986. Power seekers are different.
American Political Science Review 80 (1): 261–69.
Marcus, George E., W.R. Neuman, and Michael Mac-
Kuen. 2000. Affective Intelligence and Political Judg-
ment. Chicago: University of Chicago Press.
Martin, N.G., L.J. Eaves, A.C. Heath, R. Jardine, L.M.
Feingold, and H.J. Eysenck. 1986. Transmission of
social attitudes. Proceedings of the National Academy of
Sciences (15 June): 4364–68.
McClosky, Herbert. 1958. Conservatism and personal-
ity. American Political Science Review 52 (1): 27–45.
McDermott, Rose. 2004. The feeling of rationality:
The meaning of neuroscientific advances for
political science. Perspectives on Politics 2 (4):
691–706.
Morris-Yates, A., G. Andrews, P. Howie, and S. Hender-
son. 1990. Twins: A test of the equal environments
assumption. Acta Psychiatrica Scandinavica 81 (4):
322–6. 䡬
O’Connor, T.G., K. Deater-Deckard; D Fulker, M.
Rutter, and R. Plomin. 1998. Genotype-environment
correlations in late childhood and early adolescence:
Antisocial behavior problems and coercive parenting.
Developmental Psychology 34 (5): 970–81.
Orbell, John, Tomonori Morikawa, Jason Hartwig,
James Hanley, and Nicholas Allen. 2004. Machiavel-
lian intelligence as a basis for the evolution of cooper-
ative dispositions. American Political Science Review
98 (1): 1–16.
Plomin, R. 1990. The role of inheritance in behavior.
Science, 183–248, April 13.
Plomin, R., J.C. DeFries, G.E. McClearn, and P. McGuf-
fin. 2001. Behavioral Genetics, 4th ed. New York:
Worth.
Plomin, R., L. Willerman, and J.C. Loehlin. 1976.
Resemblance in appearance and the equal environ-
ments assumption in twin studies of personality traits.
Behavior Genetics 6: 43–52.
Scarr, Sandra, and L. Carter-Saltzman. 1979. Twin
method: Defense of a critical assumption. Behavior
Genetics 9: 527–42.
Schreiber, Darren. 2005. “Evaluating Politics: A Search
for the Neural Substrates of Political Thought.” PhD
diss., UCLA, Los Angeles, CA.
Schwartz, S. H., and A. Bardi. 2001. Value hierarchies
across cultures: Taking a similarities perspective.
Journal of Cross Cultural Psychology 32: 268–90.
June 2008 | Vol. 6/No. 2 327
| |
Exchange | Political Genotypes
Thornhill, Randy, and Corey L. Fischer. 2007. “What Is
the Relevance of Attachment and Life History to
Political Values?” Presented at the annual meeting of
the Human Behavior and Evolution Society, Wil-
liamsburg, VA, May 30–June 3.
Westen, D., P. Blagov, K. Harenski, C. Kilts, and S.
Hamann. 2006. Neural basis of motivated reasoning:
An fMRI study of emotional constraints on political
328 Perspectives on Politics
䡬
judgment during the U.S. presidential election of
2004. Journal of Cognitive Neuroscience 18: 1947–58.
Xian, H., J.F. Scherrer, S.A. Eisen, W.R. True, A.C.
Heath, J. Goldberg, M.J. Lyons, and M.T. Tsuang.
2000. Self-reported zygosity and the equal environ-
ments assumption for psychiatric disorders in the
Vietnam era twin registry. Behavior Genetics 30:
303–10.
䡬
| |
䡬

Exchange

The Threat of Genes: A Comment on

Evan Charney’s “Genes and Ideologies”
Rebecca J. Hannagan and Peter K. Hatemi

In his essay, “Genes and Ideologies,” Evan Charney wrangles with the question of the role of genes in the formation of
political attitudes via a critique of Alford, Funk, and Hibbing’s 2005 American Political Science Review article. Although critical
evaluations are necessary, his essay falls short of what is required of a scientific critique on both empirical and theoretical
grounds. We offer a comment on his essay and further contend that it is naïve to proceed on the assumption that a barrier exists
between the biological and social sciences, such that the biological sciences have nothing to offer the social sciences. If we look
beyond our discipline’s current theoretical models we may find a more thorough, and not just competing, explanation of
political behavior.

tudies examining genetic influences on behavior Methodological Considerations

S have become a significant part of scholarship repre-

sented in reputable and high impact journals across
disciplines, including Science, Nature, the Proceedings of
the National Academy of Sciences, the Journal of Economic
Literature, Psychological Bulletin, and many others. Many
of these studies focus on health issues and socially dam-
aging behaviors such as personality disorders, alcoholism,
and depression,1 but others focus on the role of genes
in social and political attitudes and behaviors.2 While
consideration of the role of genes in models of atti-
tudes and behaviors was introduced to other fields in
the 1970s, it is a relatively new addition to political sci-
ence.3 In particular, Alford, Funk and Hibbing’s (2005)
APSR article “Are Political Orientations Genetically
Transmitted?” has drawn attention to the issue of the
role of genes in political attitudes both from within
academia and the popular press. In his essay, “Genes and
Ideologies,” Evan Charney takes up the question
of the role of genes via a critique of Alford, Funk,
and Hibbing. Although a critical evaluation was long
overdue, his essay falls short of what is required of
a scientific critique on both empirical and theoretical
grounds.
Rebecca J. Hannagan is Assistant Professor, Department of
Political Science, Northern Illinois University (rhannaga@
niu.edu). Peter K. Hatemi is Post-Doctoral Research
Fellow, Virginia Institute of Psychiatric and Behavioral
Genetics, Virginia Commonwealth University (Peter.
Hatemi@qimr.edu.au).
Science is a way to ensure accountability for claims because
the veracity of results can be questioned, studies repli-
cated, and hypotheses retested using different methods.
Rather than replicating the Alford, Funk, and Hibbing 䡬
(hereafter AFH) findings using a new data set, or testing
the hypotheses employing different methods, Charney
questions the legitimacy of the AFH results through a
multi-faceted essay. Certainly, questioning the results of
an empirical study is the way in which scientific inquiry
proceeds. An adequate critique of an empirical study,
however, must entail a coherent explanation of the inad-
equacies of the methods employed and demonstrate or
suggest what methods should be used in order to better
test the hypotheses. If this is not done the evaluation is
not a scientific critique.
The methods currently employed in the fields of genet-
ics, psychiatry, and other disciplines used to explore atti-
tudes and behaviors suggest a methodological critique of
AFH is warranted. Twin studies are only a first step in
genetic epidemiological research, albeit an important one
(less frequently family or adoption studies are also used).
Classical twin studies estimate heritability (h 2 ) based on
twin correlations: h 2 is 2(rMZ ⫺ rDZ), where r is the
correlation coefficient. The relative contributions of the
shared and non-shared environmental effects are: c 2 ⫽
2rDZ ⫺ rMZ and e 2 ⫽ 1 ⫺ h 2 ⫹ c 2, respectively. According
to this formula, heritability is an estimate for the relative
contribution of genetic effects to total phenotypic vari-
ance. Following designs from earlier twin studies, AFH
applied polychoric correlations to the Holzinger formula
above.4 However, the method used by AFH is seldom
employed for raw data analyses in current scholarship.

doi:10.1017/S153759270808064X June 2008 | Vol. 6/No. 2 329

䡬
| |
Exchange | The Threat of Genes
Polychoric correlation transformations are limited in
that the formula does not (1) allow for model fitting (test-
ing whether genetic or familial influences can be removed
from the model without reducing model fit); (2) provide
confidence intervals; (3) include opposite-sex (OS) twin
pairs (thereby excluding roughly 1/3 of the sample in the
VA30K data set used by AFH); (4) test for male-female
differences in the magnitude of variance components esti-
mates; (5) test for the potential for difference in males or
females genes which influence the trait; (6) test or model
differences in either means for continuous data and/or
thresholds for ordinal data between the different zygosity
groups; and (7) allow for the modeling of age, or other
covariates. For these reasons, polychoric correlation trans-
formations have been replaced by more advanced methods.5
Structural equation modeling (SEM) under a maxi-
mum likelihood (ML) framework is the most common
method used to infer the relative importance of the cor-
relations between observed traits of monozygotic and
dizygotic twins in terms of their underlying genetic and
environmental components, less frequently, Bayesian meth-
ods are also used (see Fowler, Baker, and Dawes 2006).
SEM/ML addresses the aforementioned shortcomings by:
(1) testing for differences in the zygosity groups; (2) mod-
eling those difference if they exist; (3) including OS twins;
(4) including other siblings, parents, and any number of
䡬 different relative types; and (5) model fitting to deter-
mine if removing the genetic or social component of a
specific trait provides a statistically better model. This
last feature addresses Charney’s argument that his expla-
nation for attitudes is more parsimonious than what AFH
present. What Charney seems to be arguing is that exclu-
sively environmental explanations are “more parsimoni-
ous” simply because we are more used to them—an
outrageous assertion on the face of it. Parsimony is only
a virtue if it is also consistent with the data, not if it is
only consistent with our preconceptions. If model parsi-
mony is paramount, as Charney suggests, methodologi-
cal improvements to AFH will resolve this concern. Rather
than to assume that environment-only models are both
more parsimonious and better fitting, model fitting tech-
niques are available to determine whether a common and
unique environmental model is superior to a model includ-
ing genetic influences.
In short, the results presented by AFH can be strength-
ened and potential problems with the findings can be
addressed with more sophisticated methods currently used
in behavior genetics. An important consideration regard-
ing the use of polychoric Holzinger transformations, is
that if no sex differences exist, and the zygosity groups
present no differences in means or thresholds, then Holz-
inger transformations provide remarkably similar results
to an ML analysis.6 This happens to be the case with the
AFH results. The conservatism items in the AFH study
were tested within an ML framework using the full sam-
330 Perspectives on Politics
䡬
ple of twins.7 A subset of the results of those analyses are
shown in table 1.
While there are some sex differences in the magnitude
of the variance components, the additive genetic compo-
nent could only be dropped for one trait (political affil-
iation), while the common environment component could
be removed for more than half the attitude items. The
more sophisticated methodology and more parsimoni-
ous explanation suggest that genes play an even stronger
role in political attitudes. In other words, AFH pro-
vided a more conservative estimation using the simpler
method.
An important methodological consideration raised by
Charney concerns the construction of the conservatism
scale used by AFH. He asserts that not all political items
in the scale can be given equal weight and therefore the
analyses are flawed. This point would be better addressed
empirically by simply weighting the items based on national
polling data in the year of the study (or the appropriate
NES data) and rerunning the analyses based on an appro-
priate weighting scheme. However, a review of the AFH
findings illustrate that many items that are stronger cor-
relates to conservatism (e.g., immigration vs. divorce) have
a more pronounced genetic influence. It is likely, there-
fore, that the results from such an analysis would show an
increase in the additive genetic influence of conservatism
(e.g., Hatemi et al. 2007). This stands opposed to what
Charney seems to intend. Also, there is ample evidence
for the construct validity and heritability of the Wilson-
Patterson conservatism scale and there are two very impor-
tant considerations not raised by Charney that a brief
review of the current literature employing the Wilson-
Patterson index provides.8 Several studies factored the scale
into separate latent constructs and then ACE modeled
each factor score separately.9 In each of the factors there is
a significant genetic influence, though they differ depend-
ing on the factor (labeled Sex, Militarism, Religion, Poli-
tics, and Economics). Further, while the scale may be
limited in many respects, it is remarkably normally
distributed.
While it is no simple task to gain a broad understand-
ing of a field, or a deep understanding of a subfield, a
critique of empirical work should attempt to be thorough
in the literature it presents. A particular transgression by
Charney in this regard is his presentation of the equal
environments assumption (EEA) literature in an attempt
to augment his argument that the findings from twin stud-
ies are confounded. The literature presented was highly
selective and ignored the corrections and tests employed
by geneticists and social scientists to test the validity of the
assumption on a trait-specific basis.10
Early tests for EEA violations correlated perceived twin
similarity with the trait under consideration while con-
trolling for actual zygosity. Numerous studies of person-
ality, intelligence, and psychiatric behaviors have found
䡬
 |
Table 1
(US) Standardized Variance Components (95% CI) Sex Limitation Model Fitting for Political Attitudes; Thresholds Corrected
for Age a
Parameter Estimates
Females Males
p-value
Model a 2
c 2
e 2
a 2
c2 e2 −2LL ⌬X 2 ⌬df (comparison model)
Abortion ACE bc .26 (.12–.41) .41(.27–.53) .33 (.29–.37) .38 (.16–.51) .19 (.10–.37) .43 (.36–.50) 23249.16 6.33 6 .38 (ACE)
Astrology AE b .47(.43–.47) — .53 (.48–.57) .47 (.39–.54) — .53 (.46–.61) 24329.32 2.83 2 .24 (ACE)
Busing ACE b .31 (.16–.31) .09 (.08–.20) .60 (.55–.65) .12 (0–.40) .30 (.06–.45) .58 (.50–.66) 22772.97 3.64 4 .46 (ACE)
Capitalism AE b .47 (.43–.52) — .53 (.48–.57) .61 (.54–.67) — .39 (.33–.46) 23031.11 5.08 2 .07 (ACE)
Censorship AE b .38 (.33–.42) — .62 (.58–.67) .39 (.35–.47) — .61 (.53–.69) 24416.92 5.92 2 .05 (ACE)
Death Penalty ACE bd .35 (.22–.48) .21 (.10–.31) .44 (.40–.48) .35 (.22–.48) .21 (.10–.31) .44 (.40.-48) 18872.82 0.29 3 .96 (ACE)
Divorce ACE bc .25 (.16–.29) .23 (.08–.38) .52 (.47–.57) .42 (.31–.42) 0 (.00–.07) .57 (.53–.65) 24253.99 10.35 6 .11 (ACE)
Draft AE bd .37 (.32–.41) — .63 (.60–.68) .37 (.32–.41) — .63 (.60–.68) 22096.51 0.10 1 .75 (ACE)
Federal Housing AE b .41 (.36–.46) — .59 (.54–.64) .41 (.36–.46) — .59 (.54–.64) 22455.92 5.49 2 .06 (ACE)
Foreign Aid ACE b .40 (.29–.45) .01 (.00.-10) .59 (.55–.64) .31 (.08–.49) .11 (.00–.31) .58 (.51–.66) 25235.07 8.35 4 .08 (ACE)
䡬

䡬
Gay Rights ACE bd .34 (.24–.45) .25 (.22–.34) .41 (.39–.45) .34 (.24–.45) .25 (.22–.34) .41 (.39–.45) 22434.67 5.02 3 .17 (ACE)
Immigration AE bd .46 (.46–.49) — .54 (.51–.54) .46 (.46–.49) — .54 (.51–.54) 24832.82 1.02 1 .31 (ACE)
Living Together ACE bc .51 (.41–.68) .16 (.10–.24) .33 (.30–.37) 0 (.00–.34) .48 (.21–.54) .52 (.52–.58) 21940.29 6.82 6 .33 (ACE)
Military Drill AE bd .36 (.31–.40) — .64 (.63–.69) .36 (.31–.40) — .64 (.63–.69) 21635.07 6.88 4 .14 (ACE)
Modern Art AE bcd .40 (.36–.43) — .60 (.57–.64) .40 (.36–.43) — .60 (.57–.64) 25004.82 0.27 1 .61 (ACE)
Moral Majority AE bd .42 (.38–.47) — .58 (.53–.62) .42 (.38–.47) — .58 (.53–.62) 24882.86 1.15 2 .56 (ACE)
Nuclear Power AE bd .34 (.30–.39) — .65 (.61–.65) .34 (.30–.39) — .65 (.61–.65) 24577.93 5.99 2 .06 (ACE)
Pacifism AE bd .31 (.27–.35) — .69 (.65–.73) .31 (.27–.35) — .69 (.65–.73) 22094.26 0.79 1 .94 (ACE)
Party Affiliation CE bcd — .81 (.78–.84) .19 (.16–.22) — .81 (.78–.84) .19 (.16–.22) 8738.75 2.34 2 .31 (ACE)
Property Tax AE bd .42 (.41–.46) — .58 (.58–.63) .42 (.41–.46) — .58 (.58–.63) 21227.90 0.00 1 .48 (ACE)
Religiosity-2 ACE bc .56 (.35–.66) .19 (.08–.39) .25 (.21–.29) .22 (.00–.57) .36 (.05–.59) .41 (.32–.50) 15047.54 3.33 3 .34 (ACE)
School Prayer ACE b .32 (.16–.48) .37 (.22–.51) .31 (.27–.36) .47 (.22–.62) .21 (.09–.41) .32 (.26–.40) 18018.47 4.66 4 .32 (ACE)
Segregation AE bcd .37 (.32–.37) — .63 (.59–.68) .37 (.32–.37) — .63 (.59–.68) 20367.82 0.08 1 .78 (ACE)
June 2008 | Vol. 6/No. 2 331

Socialism AE bd .38 (.34–.38) — .62 (.58–.66) .38 (.34–.38) — .62 (.58–.66) 21328.12 0.53 1 .46 (ACE)
Unions AE bd .41 (.36–.46) — .59 (.54–.64) .41 (.36–.46) — .59 (.54–.64) 24884.86 4.34 2 .11 (ACE)
Women’s Lib ACE bc .34 (.18–.49) .18 (.05–.18) .48 (.44–.53) .31 (.23–.39) 0 (.00–.03) .69 (.61–.76) 24217.86 8.22 6 .22 (ACE)
X-Rated Movies AE bcd .51 (.47–.56) — .49 (.46–.54) .51 (.47–.56) — .49 (.46–.54) 18652.25 0.79 2 .67 (ACE)
Note: This table was originally published in Hatemi 2007 as table 4.3; (a) Only best fitting models shown (b) Equated Thresholds for MZ and DZ pairs (MZ/DZ groups have no
difference).(c) Equated Thresholds for Males and Females (d) Equated Variance components for Males and Females.
䡬

|
| |
Exchange | The Threat of Genes
that twin trait resemblance was not influenced by per-
ceived similarity.11 Later tests focused on correlating the
similarity of the twins’ environments with the trait under
consideration while controlling for actual zygosity, but
also found no violation of the EEA.12 A more recent test
modeled the discrepancy between perceived and actual
zygosity by extending the ACE model to partition the
common environment into two parts; usual common envi-
ronment (which is correlated at 1.0 for all twin pairs) and
specific common environment (which is determined by
perceived zygosity—correlated at 1 if both twins perceive
themselves to be MZ, 0 if both twins see themselves as
DZ, and .5 if the twins disagree about their zygosity).
Utilizing this method, studies find no evidence that per-
ceived zygosity (whether from the twins, parents, or oth-
ers) influenced resemblance for personality traits or social
attitudes.
The most common and least prohibitive test currently
used for ordinal data is a simple statistical comparison
which tests if equating the prevalence of the trait or atti-
tude under examination (modeled as thresholds within a
multi-factorial threshold model) between MZ and DZ
twins provides a better fit to the data than separate thresh-
olds. Thresholds for MZ and DZ twin pairs that can be
equated without worsening model fit implies no differ-
ence in variances between MZ and DZ twin pairs.13 Dif-
䡬 ferences have been found for traits such as perceived
closeness to siblings but not for intelligence, personality,
or social and political attitudes.
Potential violations of the EEA are important to recog-
nize when critiquing a twin study, but it should be noted
that the EEA must be tested for each specific trait under
consideration. For example, if a violation is found for
dressing alike it does not follow that the violation applies
to height, weight, or political attitudes. While a violation
of the EEA invalidates the use of the classical twin model
for the specific trait in question, it does not invalidate it
for every trait. Social science challenges to the EEA tend
to either misrepresent the assumption or generalize
violations—suggesting that a violation means the overall
twin design is empirically unsound and its results cannot
be trusted for any trait. Again, a systematic review of the
literature reveals these important considerations.
The EEA does not mean that geneticists assume there
are no differences in MZ and DZ rearing. Rather, they
assume that these differences do not affect the trait under
examination because they test for it and if a violation is
found they model the violation to correct for it. In the
rare instances that a violation may be present, a statistical
technique is used to model the increased similarity rather
than assuming that similarity in dress, classrooms, and
room sharing make any difference in twins’ similarity in
political attitudes.
The data and methods employed by AFH in their 2005
American Political Science Review article were presented
332 Perspectives on Politics
䡬
for critique by their peers in the scientific tradition. If we
collectively aim at becoming a science of human behavior,
we have an obligation to hold each other accountable within
scientific standards. Although critiquing AFH is entirely
appropriate, the critique offered by Charney provides no
opportunity to hold AFH accountable for their findings.
The argument can be made that AFH’s results should be
carefully scrutinized because of the method used to test
their hypotheses—similar to what we suggest above—but
while AFH used simpler methods than those employed in
current research, it should be noted that most studies
attempting to analyze the heritability of any phenotype
start with simple correlation differences between MZ and
DZ twins. Normal science proceeds incrementally and
complex studies such as those undertaken by behavioral
geneticists are typically executed in stages.
Theoretical Considerations
It is widely acknowledged that attitudes are learned, that
is, that they develop through experience.14 It is often
argued, in light of this understanding, that attitudes are
environmentally caused. Studies since the 1970s have
reported modest to strong genetic influences on social and
political attitudes, thereby providing empirical evidence
that attitudes and behaviors are a result of both genes and
environment.15 Such findings do not negate the impact of
the environment, but explain the extent to which environ-
ment matters. Regarding the possibility of genetic influ-
ence on political attitudes, Charney says that “such a
hypothesis is, in and of itself, extremely implausible (if
not incoherent).” Addressing this admittedly widespread
belief among social scientists appears to be the goal of the
AFH study. However, it is not the case that environmental
and biological hypotheses are implausible, incoherent, or
incompatible—in fact, they are inseparable. Genetic fac-
tors exert their influence on an organism in a particular
environment such that any trait must be a combination of
the two factors. Any explanation that denies this inter-
action is incoherent.
Regarding the role of genes, Charney states that
if true, it would require nothing less than a revision of our under-
standing of all of human history, much, if not most of political
science, sociology, anthropology, and psychology, as well as, per-
haps, our understanding of what it means to be human.16
This “revision of our understanding” happened a very long
time ago. There is no nature-nurture debate. One can
review the classical literature from Darwin (1872) and his
contemporaries, or more recent literature from genetics to
neuroscience and even philosophy to discern that the sci-
entific community recognizes that genes are very much a
part of what it means to be human. Genes are not myste-
rious, elusive or fleeting, and any assertion to the contrary
is comparable to suggesting that bacteria we cannot see
are not really there and that evil spirits in the body cause
䡬
| |
䡬
illness. Genes are physical and quantifiable. With empir-
ical study, increasingly sophisticated methodology,
technology, and time, scientists will continue to under-
stand how genes do what they do.
Charney’s critique seems largely a philosophical strug-
gle with empirical science and the “threat” of genes merely
provides the impetus. The argument he attempts to put
forth is that variance component estimates do not mea-
sure what AFH say they measure. Charney does not con-
vincingly show this to be the case. He may not believe
such methods measure what they purport to measure, but
science does not proceed in this way. If the problem is
ultimately a disbelief in the ability to measure human
attitudes and behavior, we relegate our discipline to phi-
losophy and history. Charney’s use of the Horwitz et al.
(2003) critique of the EEA illustrates this danger. Hor-
witz et al.’s (2003, 125) assertion that “theoretical assump-
tions not empirical findings determine where to end the
chain of causation between social and genetic factors” is
an attempt to critique empirical findings on moralistic
grounds.17 A more scientifically-oriented critique would
offer a means for further clarifying the estimates produced
through twin studies rather than dismissing them simply
because they only provide estimates. Estimates from any
empirical study, whether regression analysis, Bayesian mod-
els, or others, are all just estimates based on the model
employed.
Conclusion
It is unlikely that “the” gene for conservatism, financial
success, a great golf stroke, or any other complex trait will
be identified. It is more likely that complex networks of
genes, for which causal variation might be specified, are
the appropriate targets for future research. Genes likely
establish general inclinations or predispositions that shape
our interpretation and reaction to experiences. Those expe-
riences increase the likelihood of developing a specific trait
or attitude.18 It may be the case that the more we learn
about genes the more we discover the importance of rele-
vant environmental influences on behavior. Without con-
sideration of one we would not gain full understanding of
the other.
If what political scientists are truly after is an answer to
the question, “Why do people do what they do?” a focus
on cultural or social influences alone will leave us with an
incomplete understanding of our subject. Social determin-
ism does not make any more sense than biological or genetic
determinism and to proceed on the assumption that a
barrier exists between the biological and social sciences,
such that the biological sciences have nothing to offer the
social sciences is naïve.
This comment is not intended to be simply an exami-
nation of one author’s misrepresentative attack on a par-
ticular study, but a response to the idea that is it acceptable
䡬
to critique empirical work based on the philosophical rejec-
tion of the scientific method. Epidemiologists and psy-
chologists are currently undertaking the study of political
attitudes and behaviors. We may prefer to wrangle about
the utility and philosophical implications of incorporat-
ing genes into our models, but the evidence suggests that
we must take on new theoretical approaches and develop-
ments in methodology and consider them as candidate
improvements upon our existing paradigm. The alterna-
tive is to yield significant parts of our discipline to scien-
tists in other disciplines. To concur with Charney, the
AFH study could doubtless be improved upon—but what
scientific study cannot? The greater issue emerging from
this critique is that if we are to proceed as a social science
there is something to learn from AFH. If we look beyond
our discipline’s current theoretical models we may find a
more thorough, and not just competing, explanation of
political behavior.
Notes
1 Caspi et al. 2002, 2003.
2 Eaves and Eysenck 1974; Eaves et al. 1989; Martin
et al. 1986; Truett et al. 1992.
3 Alford, Funk, and Hibbing 2005; Fowler 2006,
2007; Hatemi, Alford, Hibbing, Keller, Martin,
Medland, and Eaves 2007; Hatemi et al. 2007b; 䡬
Fowler and Dawes 2007; for an exception see Nel-
son 1974.
4 Holzinger 1929.
5 Rijsdijk and Sham 2002; and Neale 1997, 2000.
6 Neale and Cardon 1992.
7 Hatemi 2007.
8 Bouchard et al. 2003.
9 Eaves et al. 1999.
10 Matheny, Wilson, and Dolan 1976; Plomin and
Lachlin 1976; Scarr and Carter-Saltzman 1979.
Also, Lytton 1977 examined family members and
found no relationship between the parent’s percep-
tion of the twin’s zygosity and actual twin behavior.
11 Kendler 1983; Loehlin and Nichols 1976; Kendler
et al. 1987; Martin et al. 1986; Heath, Jardine, and
Martin 1989.
12 Hettema, Neale, and Kendler 1995; Kendler et al.
1993; Xian et al. 2000. Hettema, Neale, and Ken-
dler did find an equal environment assumption
violation for Bulimia Nervosa. No violations were
found for other psychological traits (e.g., major
depression, generalized anxiety disorder, phobia, and
alcoholism).
13 Many authors describe this test more generally stat-
ing they are testing for “twin specific effects” and do
not explicitly state that they are testing for the EEA.
In addition, when data have been collected from
non-twin siblings, checking for differences in the
June 2008 | Vol. 6/No. 2 333
| |
Exchange | The Threat of Genes
variances or thresholds between twins and siblings,
and for differences between the DZ covariance and
the twin-sibling and sibling-sibling covariances, can
provide a more robust test of the EEA.
14 Eagly and Chaiken 1993.
15 Eaves and Eysenck 1974; Eaves et al. 1989; Martin
et al. 1986; Truett et al. 1992.
16 Charney 2008, 330.
17 Horwitz et al. 2003; but see Freese and Powell 2003
for an overview of this debate.
18 Olson et al. 2001.
References
Alford, John, Carolyn Funk, and John R. Hibbing. 2005.
Are political orientations genetically transmitted?
American Political Science Review 99 (2): 153–168.
Bouchard, Thomas J., Nancy L. Segal, Auke Tellegen,
Matt McGue, Margaret Keyes, and Robert Krueger.
2003. Evidence for the construct validity and herita-
bility of the Wilson-Patterson conservatism scale: a
reared-apart twins study of social attitudes. Personality
and Individual Differences 34: 959–69.
Caspi, Avshalom, Joseph McClay, Terrie Moffitt, Jonathan
Mill, Judie Martin, Ian Craig, Allen Taylor and Richie
Poulton. 2002. Role of genotype in the cycle of vio-
䡬 lence in maltreated children. Science 297: 851–53.
Caspi, Avshalom, Karen Sugden, Terrie Moffitt, Alan
Taylor, Ian Craig, Honalee Harrington Joseph Mc-
Clay, Jonathan Mill, Judie Martin, Antony Braith-
waite, and Richie Poulton. 2003. Influence of life
stress on depression. Science 301: 386–89.
Darwin, Charles. 1998 [1872]. The Expression of Emo-
tions in Man and Animals, ed. Paul Eckman. Oxford:
Oxford University Press.
Eagly, A.H., and S. Chaiken. 1993. The Psychology of
Attitudes. San Diego, CA: Harcourt Brace Jovanovich.
Eaves, L.J., and H.J. Eysenck. 1974. Genetics and the
development of social attitudes. Nature 249: 288–89.
Eaves, L.J., H.J. Eysenck, and N.G. Martin. 1989.
Genes, Culture and Personality: An Empirical Approach.
New York: Academic Press.
Eaves, Lindon, Andrew Heath, Nicholas Martin,
Hermine Maes, Michael Neale, Kenneth Kendler,
Katherine Kirk, and Linda Corey. 1999. Comparing
the biological and cultural inheritance of personality
and social attitudes in the Virginia 30,000 study
of twins and their relatives. Twin Research 2: 62–80.
Fowler, James. 2006. “The Genetic Basis of Political
Cooperation.” Presented at the annual meeting of the
American Political Science Association, Philadelphia,
PA. August 30–September 2.
_. 2007. “The Genetic Basis of Voter Turnout.” Pre-
sented at the annual meeting of the Midwest Political
Science Association, Chicago, IL. April 12–15.
334 Perspectives on Politics
䡬
Fowler, James, Laura Baker, and Christopher Dawes.
2006. “The Genetic Basis of Politicl Cooperation.”
Presented at the Hendricks Conference on Biology
and Politics, Lincoln, NE, October 13–14.
Fowler, James, and Chris Dawes. 2007. “The Genetic
Basis of Political Participation.” Presented at the
annual meeting of the American Political Science
Association, Chicago, IL, August 30–September 2.
Freese, Jeremy, and Brian Powell. 2003. Tilting at
twindmills: Rethinking sociological responses to
behavioral genetics. Journal of Health and Social Be-
havior 44 (2): 130–35.
Hatemi, Peter K. 2007. “The Genetics of Political Atti-
tudes.” PhD diss., University of Nebraska-Lincoln.
Hatemi, Peter, John Alford, John Hibbing, Matthew
Keller, Nicholas Martin, Sarah Medland, and Lin-
don Eaves. 2007. “Not by Twins Alone: Using
the Extended Twin Family Design to Investigate
the Genetic Basis of Political Beliefs.” Presented
at the annual meeting of the American Political
Science Association, Chicago, IL, August 30–
September 2.
Hatemi, Peter K., Sarah E. Medland, Katherine I. Mor-
ley, Andrew C. Heath, and Nicholas G. Martin.
2007. The genetics of voting: An Australian twin
study. Behavior Genetics 37 (3): 435–48.
Heath, A.C., R. Jardine, and N.G. Martin. 1989. Inter-
active effects of genotype and social environment on
alcohol consumption in female twins. Journal of
Studies on Alcohol 60: 38–48.
Hettema, J.M., M.C. Neale, and K.S. Kendler. 1995.
Physical similarity and the equal-environments as-
sumption in twin studies of psychiatric disorders.
Behavior Genetics 25: 327–35.
Horwitz, Allan, Tami Videon, Mark Schmitz, and Diane
Davis. 2003. Rethinking twins and environments:
Possible social structures for assumed genetic influ-
ences in twin research. Journal of Health and Social
Behavior 44 (2): 111–29.
Kendler, K.S. 1983. Overview: Current perspective on
twin studies of schizophrenia. American Journal of
Psychiatry 140: 1413–25.
Kendler, K.S., A.C. Heath, N.G. Martin, and L.J.
Eaves. 1987. Symptoms of anxiety and symptoms of
depression. Same genes, different environments?
Archives of General Psychiatry 44: 451–57.
Kendler, K.S., M.C. Neale, R.C. Kessler, A.C. Heath,
and L.J. Eaves. 1993. A test of the equal-environment
assumption in twin studies of psychiatric illness.
Behavior Genetics 23: 21–27.
Loehlin, J.C., and R.C. Nichols. 1976. Heredity, Envi-
ronment, and Personality: A Study of 850 Sets of Twins.
Austin, TX: University of Texas Press.
Lytton, H. 1977. Do parents create, or respond to, differ-
ences in twins? Developmental Psychology 13: 456–59.
䡬
| |
䡬
Martin, N.G., L.J. Eaves, A.C. Heath, R. Jardine, L.M.
Feingold, and H.J. Eysenck. 1986. Transmission of
social attitudes. Proceedings of the National Academy of
Science 83: 4364–68.
Matheny, A.P., R.S. Wilson, and A.B. Dolan. 1976.
Relations between twins’ similarity of appearance and
behavioral similarity: Testing an assumption. Behavior
Genetics 6: 343–51.
Neale, M.C. 1997. Mx: Statistical Modeling (Box
980126) 3d ed. Richmond, VA: MCV.
_. 2000. QTL mapping with sib-pairs: The flexibil-
ity of Mx. In Advances in Twin and Sib-Pair Analysis,
ed. T.D. Spector, H. Snieder, and A.J. MacGregor.
London: Oxford University Press.
Neale, M.C., and L.R. Cardon. 1992. Methodology for
Genetic Studies of Twins and Families. Dordrecht, The
Netherlands: Kluwer Academic Publishers.
Nelson, Stephen D. 1974. Nature/nurture revisited I: A
review of the biological bases of conflict. Journal of
Conflict Resolution 18 (2): 285–335.
Olson, James M., Philip A. Vernon, Julie Aitken Harris,
and Kerry L. Jang. 2001. The heritability of attitudes:
䡬
A study of twins. Journal of Personality and Social
Psychology 80 (6): 845–60.
Plomin, R., L. Willerman, and J.C. Loehlin. 1976.
Resemblance in appearance and the equal environ-
ments assumption in twin studies of personality traits.
Behavior Genetics 6: 43–52.
Rijsdijk, Frühling V., and Pak C. Sham. 2002. Analytic
approaches to twin data using structural equation
models. Briefings in Bioinformatics 3 (2): 119–33.
Scarr, S., and L. Carter-Saltzman. 1979. Twin method:
Defense of a critical assumption. Behavior Genetics 9:
527–42.
Truett, K.R., L.J. Eaves, J.M. Meyer, A.C. Heath, and
N.G. Martin. 1992. Religion and education as medi-
ators of attitudes: A multivariate analysis. Behavior
Genetics 22: 43–6.
Xian, Hong, Jeffrey F. Scherrer, Seth A. Eisen, William
R. True, Andrew C. Heath, Jack Goldberg, Michael J.
Lyon, and Ming T. Tsuang. 2000. Self-reported zygos-
ity and the equal-environments assumption for psy-
chiatric disorders in the Vietnam era twin registry.
Behavior Genetics 30: 303–10.
June 2008 | Vol. 6/No. 2 335
| |
䡬
Exchange
Politics, Genetics, and “Greedy 1
Reductionism”
Evan Charney
would like to thank Alford, Funk, and Hibbing, and
I Hannagan and Hatemi, for agreeing to write critical
responses to my article, and I am grateful for the oppor-
tunity afforded me to respond.
I begin with the response of Alford, Funk, and Hib-
bing. An enforced limitation on space requires some selec-
tivity in my response, but I have tried to respond to all of
their major critiques.
Alford, Funk, and Hibbing (AFH) begin by implying
that I am caught in the “hubris” of a “mind-body dual-
ism” which sees the mind as some kind of mystical entity.
Nowhere in my article have I said anything at all regard-
ing “mind-body dualism.” It is not an assumption I argue
for, nor is it a premise of any of the arguments I make.
Perhaps AFH believe that if genes are not accepted as the
ultimate causal explanation for any and all human phe-
nomena, then somehow this implies “mind-body dual-
ism.” It does not.
AFH claim that I err in treating the environment as
“exogenous”: “Whether the causal order is [genes r beliefs]
or [genes r physical traits r social reactions r beliefs], the
underlying cause is still genetic.” To illustrate this point,
AFH assert that “negative parenting is typically assumed
to be the cause of children’s antisocial behavior, when in
point of fact children play an important role in shaping
their own environment, in this case by influencing the
behavior of their parents.” 2 What AFH are arguing is that
children, presumably on the basis of genetic traits, create
or elicit negative parenting, and the effects on children of
this negative response on the part of parents should itself
be counted as “genetic” (i.e., caused by the genetic traits
of the child in question).
This tendency to view gene-environment covariance as
a genetic effect is not uncommon among psychologists
who undertake twin studies. Thus, according to Bou-
chard, “[identical] twins tend to elicit, select, seek out, or
create very similar effective environments and, to that
extent, the impact of these experiences is counted as a
doi:10.1017/S1537592708080651
䡬
genetic influence.” 3 Suppose that a one-year-old child is
cranky and is beaten by its parents. According to Bou-
chard and AFH, the impact upon the child of its being
beaten by its parents is itself genetic because, presumably,
the fact that the child is cranky is genetic. But of course,
not all parents are abusive and beat their children in
response to a child’s cranky behavior (genetic or other-
wise), and for any given pair of children with similar dis-
positions, parental responses will be as varied as parenting
styles.
But the absolutely fallacious nature of the assumption
that the effects of behavior which is a response to a “genetic
trait” should itself be counted as genetic can be seen by 䡬
considering that slavery of blacks was the response of a
group of individuals (white Europeans and Americans) to
a genetically transmitted trait, i.e., black skin color. Are
we to assume then, that the effects upon blacks of their
enslavement by European whites were genetic, because slav-
ery was “caused” or “elicited” or “created” by the genetic
trait of black skin color? Rape is a response of some men
to the genetic characteristic of being female. Should we
say that the effects upon women of being raped are genetic?
If AFH wish to distance themselves from such prepos-
terous (and pernicious) conclusions, then they are going
to have to distinguish between two types of “impacts”
upon individuals from behavior that is deemed a response
to a genetic trait: Impacts that will be deemed genetic and
impacts that will not be deemed genetic. On what basis
might they propose to draw such a distinction?
In support of the validity of the Equal Environment
Assumption (EEA), AFH reference studies of so-called
“reverse zygosity.” These studies, as noted in my article,
concern a tiny subpopulation of DZ twins mistakenly
thought by their parents to be MZ twins, and purport to
show that the degree of correspondence between MZ twins
still exceeds that of DZ twins. According to AFH, “the
results of these mis-categorization studies are clear: DZ
twin pairs believed by their environments to be MZ twin
pairs are no more similar than DZ twin pairs believed to
June 2008 | Vol. 6/No. 2 337
| |
Exchange | Politics, Genetics, and “Greedy Reductionism”
be DZ twin pairs.” But the results of such studies are not
clear. To repeat what I said in my article:
Most of these studies of “reverse zygosity,” however, relied upon
parental accounts of how they raised their twin children many
years later. Because of problems with biased impressions, poor
memory, and poor reliability, studies that rely upon parental
recall of their child rearing practices have been shown to be
notoriously unreliable, typically showing reliability measures of
only 0.3–0.5.4
In addition, studies have shown that parents’ recounts of
their rearing practices are often biased to match some ideal
of parenting.
How could an assumption as momentous (and coun-
terintuitive) as the EEA rest upon such shaky ground? Are
we to assume, on the basis of the potentially biased and
faulty memories of elderly parents concerning their child
rearing practices, that the debate concerning the validity
of the EEA has been solved once and for all? According to
what standard of science are we to accept this as conclusive
evidence of anything? 5
I now consider what AFH characterize as “the most
important challenge we wish to issue to Charney concerning
the EEA”:
We estimate the heritability of political and social attitudes to be
in the .4 to .5 range, leaving .5 to .6 attributable to environmen-
䡬 tal factors. But these same procedures reveal that party identifi-
cation is only about .14 heritable, leaving .86 attributable to the
environment, so the classical twin design reports a dramatic dif-
ference in the heritability of political beliefs and party identifi-
cation”. . . If violations of the EEA are responsible for reported
heritability, Charney must argue that parents of MZ and DZ
twins raise their children equally similarly with regard to party
identification but differentially with regard to political attitudes.6
In response, I focus on the most obvious problem with
the “challenge” AFH set before me, or rather, with the
obvious answer. AFH appear to treat “party identifica-
tion” as a fixed variable, such that we can say of any given
individual that she possesses political beliefs (“ideology”)
A and party identification B, and both are supposed to be
invariable.
Let me highlight two well-known points concerning
party identification in the United States: First, party iden-
tification in the United States is relatively weak compared
to a number of other countries, as indicated by the inci-
dence of ticket-splitting (in the 2000 elections, 20 percent
of voters split their ballots by voting for candidates from
different parties for president and for the U.S. House of
Representatives); the existence of a sizable number of vot-
ers who consider themselves Independents (and hence
aligned with neither party); and the frequency with which
Americans change political parties (a recent Pew poll
showed a sharp change in Americans’ political party iden-
tification: Democrats now outnumber Republicans 50 per-
cent to 35 percent, as opposed to 2002, when both had 43
338 Perspectives on Politics
䡬
percent). In one study, Brody and Rothenberg showed
that fewer than half the 1980 voters were stable through-
out their campaign year in their self-description on the
National Election Study party identification question.7 In
fact, party identification is so variable in the United States
that shifts in party identification can be correlated with
specific political events (see figure 1). Second, political
parties in the United States exhibit relatively low internal
unity and lack strict adherence to an ideology or set of
policy goals, allowing for a less strict alignment between
ideology and party identification.8
Hence, the findings of AFH—high ideological corre-
lation and low party identification correlation—are exactly
what one would expect given the nature of party identifica-
tion in the United States. Given the relatively weak and
variable nature of party identification, we would expect
political ideology (whatever its origins) to be more endur-
ing, more fixed and constant, than party identification
(which might differ depending upon when any given
individual is asked to identify her party affiliation). MZ
twins are more likely to share political attitudes as opposed
to party identification because of the relatively weak and
highly variable nature of the correlation between political
attitudes and party identification in the United States.
Parents’ identities are defined much more strongly by
their political ideologies (connected, as they generally are,
with their moral and religious world views) than by their
party affiliations, and for this very reason we would expect
parents to be much more concerned to transmit their
political ideologies, rather than their party affiliations, to
their children. For one committed to an “environmental”
explanation of political ideology, what AFH present as
the “ultimate challenge” ironically appears to affirm the
“environmental” approach.
AFH claim that my referencing a study by Cooper and
Zubek 9 that indicates the manner in which heritability of
maze-running ability in mice can vary dramatically with
different environments somehow undercuts my argu-
ment. First, they draw attention to the fact that I admit
that the mice in the study “may well have inherited what-
ever genes are linked to intelligence.” I have absolutely no
difficulty in acknowledging such a claim. My argument
does not concern the question of the heritability of mouse
intelligence, nor does it concern the heritability of human
intelligence. My argument concerns the intelligibility of
the proposition that political ideologies are heritable, a prop-
osition radically different than the proposition that intel-
ligence is (partially) heritable.
Second, my point in mentioning the mouse study was
not to warn against overestimating the role of genetics in
a given environment (as AFH interpret it), but simply
to emphasize the point that high heritability does not
imply that a given trait is resistant to environmental influ-
ence, contrary to what AFH say in their article, and that
䡬
| |
䡬

Figure 1
2001–2004 party ID quarterly averages with key events

Note: All data from PSRA/Newsweek polls with the following exceptions: Second quarter 2001 includes data from Pew’s June News
Interest Index (6/13–17/01). Third quarter 2001 includes data from Pew’s July Favorability poll (7/2–12/01) and Kaiser’s August
Health News Index (8/2–5/01). Fourth quarter 2002 includes data from Kaiser’s December Health News Index (12/6–10/02). Third
quarter 2004 includes data from Pew’s August Convention (8/5–10/04) and Kaiser’s August Health Poll Report (8/5–8/04).

heritability itself can change dramatically with changes in
the environment. According to AFH:
Inherited attitudes seem to be demonstrably different than
acquired attitudes. . . [A]ttitudes higher in heritability are man-
ifested more quickly, are more resistant to change, and increase
the likelihood that people will be attracted to those who share
those particular attitudes. . . To the extent that political ideolo-
gies are inherited and not learned they become more difficult to
manipulate.10
Claims of this sort are indicative of a popular misun-
derstanding of the concept of heritability: High heritabil-
ity does not mean inevitability of phenotypic outcome. 11
In response to my contention that it seems unusual that
those issues that one would associate most with social
conservatism in the United States at present—abortion
and gay rights—show lower levels of “heritability” than
other issues, AFH respond with what can only be described
as a deus ex machina: “Assortative mating,” based on the
observation that conservatives tend to marry conserva-
tives and liberals tend to marry liberals. Then, proceeding
on the assumption that parents are genetically similar, AFH
“correct” for assortative mating, and note that with this
“correction,” abortion and gay rights come out among the
top five “heritable” issues. What precisely is this supposed
to show? Is not the assumption that parents are genetically
similar something that must be proved rather than assumed
(particularly when such an assumption allows one to
manipulate the data in a manner that is more favorable to
one’s hypothesis)? When discussing “assortative mating,”
AFH comment, “while some of this interspousal similar-
ity [as determined by their answers to the 28 items on the 䡬
Wilson-Patterson index] could plausibly be attributed to
persuasion effects taking place after mate choice rather
than assortative mating, the levels of similarity are proba-
bly too high to dismiss assortative mating entirely.” 12 “Prob-
ably too high” according to what standard? Once again,
one must ask what standard of scientific evidence is being
employed here?
According to AFH, I am taken with the “context bound”
nature of words, and argue that given that words like “lib-
eralism” have no meaning for much of the world, liberal-
ism cannot be genetic. This is a parody of my argument,
which has nothing to do with words, but rather with con-
cepts (or “attitudes”), and more specifically, the clusters of
complex concepts that comprise a political ideology (and
what these clusters of concepts are named makes little dif-
ference). When AFH assert, erroneously, that “the package
of attitudes held, for example, by conservatives in the United
States is remarkably similar to that held by conservatives in
other cultures and at earlier times in American history,” 13
they discuss precisely what I discuss. Perhaps they believe,
in line with their claims concerning the trans-cultural and
trans-historical nature of liberalism and conservatism, that
all that has ever changed historically regarding these two
ideologies (from their origins to the present) is their names.
AFH assert that “no scientifically literate person in this
day and age can claim that genes are irrelevant to human

June 2008 | Vol. 6/No. 2 339

䡬
| |
Exchange | Politics, Genetics, and “Greedy Reductionism”
behavior and predispositions.” 14 Limiting what I say to
behavior, AFH are absolutely right, and nowhere do I make
such a preposterous claim. But the problem is that as dif-
fuse a term as “behavior” could cover everything from
manual dexterity and the use of language to a belief in the
doctrine of the Trinity and preferring the Yankees to the
New York Mets. My argument does not concern the ques-
tion of the “heritability” of everything that could possibly
fall under the rubric of “human behavior”; it concerns the
assumption that specific political ideologies could be genet-
ically transmitted.
The quoted sentence continues as follows: “yet many
people are deeply discomfited by this reality” (similarly,
Hannagan and Hatemi assert I am “threatened” by genetic
explanations of political ideologies). I find the ascription
of psychological motives to me by AFH (as well as Han-
nagan and Hatemi) an extremely tedious (and somewhat
adolescent) form of ad hominem arguing. It is easy to play
at such a silly game; e.g., many authoritarian personality
types have an overwhelming fear of lack of uniformity,
multiplicity of explanations, “contextualism” (the need to
consider particular cultural and historical contexts) and
“irreducibility,” and hence are drawn to simplistic, reduc-
tionist, “absolutist” explanations in which all human phe-
nomena can be reduced to a single, uniform, explanatory
variable (e.g., genes, self-interest, rational choice, “eco-
䡬 nomic rationality,” class struggle, God’s plan for human-
kind). I ascribe no such underlying psychological motives
to AFH or to Hannagan and Hatemi, and I would appre-
ciate it if they would return the favor.
According to AFH,
casting the issue as genes competing with the environment, as
Charney does in his conclusion (“if genes count for more than
environment the phenomena of liberalism and conservatism . . .
become utterly incomprehensible”) is silly and misses the point.
What we claim is that genes are important to political thought
and behavior.15
But all that I was doing was summarizing the authors’
own assertions, e.g.,
setting aside the important special case of party identification,
we find that political attitudes are influenced much more heavily
by genetics than by parental socialization. For the overall index
of political conservatism, genetics accounts for approximately
half of the variance in ideology, while shared environment includ-
ing parental influence accounts for only 11%. And in the case of
the variance in people’s tendencies to possess political opinions
at all, regardless of their ideological direction, genetics explains
one-third of the variance, and shared environment is completely
inconsequential.” 16
If AFH find such assertions silly, then we are in complete
agreement.
Finally, AFH (and Hannagan and Hatemi) are at great
pains to portray me as someone without any real under-
340 Perspectives on Politics
䡬
standing of advances in modern genetics. I will simply
point out that, as noted in my article, the section that
deals specifically with genetics and the methodology of
twin studies presents arguments that are not my own (I
cannot claim credit for them, as much as I would like to).
Rather, they are the current arguments of some of the
most prominent living geneticists, neuroscientists, and
medical researchers (some of whom have provided invalu-
able guidance), individuals such as Richard Lewontin,
Douglas Wahlstein, Jonathan Beckwith, and Annette
Karmiloff-Smith.
This “celebrity appeal” is not intended to resolve any
controversies, but it is intended to highlight the follow-
ing: The strongest critiques of the methodology of twin
studies (as well as the understanding of heritability on
which they rely) at the present time come from promi-
nent geneticists, biologists, neurologists, and medical
researchers—not, obviously, from political scientists, and
not from psychologists (with a few notable exceptions).
There exists no consensus on these matters among scien-
tific experts in genetics engaged in cutting-edge research
(among whom neither I, nor Alford, nor Funk, nor Hib-
bing, nor Hannagan, nor Hatemi can be counted). Any
claim to the contrary is manifestly false, and if I succeed
in conveying nothing more than this reality to the polit-
ical science community I will be satisfied that I have
accomplished a great deal.
A strict limitation on space will not allow a detailed
response to the comments of Hannagan and Hatemi, but
inasmuch as they make many of the same claims as AFH
in their response, I shall limit myself to two comments.
Hannagan and Hatemi (HH) place much emphasis
upon the statistical technique of structural equation mod-
eling (SEM). Advances in statistical methodology can bring
with them significant advances in scientific understand-
ing, but what they cannot do is transform a foundation-
ally flawed empirical research technique into a sound one.
Let me emphasize the following: As noted in my article,
twin studies that employ SEM rely every bit as much as
older twin studies—and the study of AFH—on the equal
environment assumption, unbiased samples, and accurate
measurements of the phenomenon being studied. None
of the objections raised to the EEA are answered, obvi-
ated, or rendered moot by SEM.
The second objection of HH that I would like to address
I consider much more interesting: It is that my objections
to AFH are not “scientific,” but rather “philosophical,”
and that I illegitimately “critique an empirical work based
on the philosophical rejection of the scientific method.”
To point out the flaws in a supposedly scientific method-
ology (twin studies), to point out that it fails to meet the
rigorous criteria of scientific knowledge, is hardly to reject
䡬
| |
䡬
the scientific method, but rather to uphold it. I do not
believe that what AFH and HH are doing is science, for
all the trappings of science they employ, e.g., an empirical
study, the collection of data, analysis of the data using
statistical methods. I could perform the exact same study
as AFH using a different questionnaire and claim to have
determined what percentage of an individual’s belief con-
cerning the doctrine of the Trinity is due to genes and
what percentage to environment—or to what extent
whether one favors the New York Yankees or the Boston
Redsocks, or Mercedes or BMWs, or Lowes or Home
Depot—is “heritable.”
A twin study which asked MZ and DZ twins whether
they shopped at Macy’s department store would very likely
reveal that this “trait” was partially “heritable” (because
MZ twins have greater contact throughout life than DZ
twins and tend to live closer to one another, they are
more likely to shop at the same stores than DZ twins).
Researchers, blind to the more obvious explanations for
their findings, might propose a “Macy’s gene.” More
sophisticated researchers might propose a Macy’s person-
ality type, correlated with the phenotype of shopping at
Macy’s. The results of such a study would doubtless
generate spectacular news headlines (and be a godsend
as a marketing tool), but if taken seriously would indi-
cate nothing more than the inadequacy and crudeness of
the researchers’ methodology (as well as their general
thinking—or lack thereof—about the plausibility of such
things). But such a study would not constitute an addi-
tion to—or advance in—scientific understanding.
Parts of my argument dealt with something called “com-
mon sense” (and it makes little difference whether or not
one wants to call this “philosophy”). Let us recall that in
their article, AFH talk of specific genes for each of the
specific beliefs that they associate with the liberal and con-
servative (or “contextualist” and “absolutist”) phenotypes:
Even if the individual genes involved with absolutism or contex-
tualism tend to move together, this does not mean they always
do. Some individuals may carry, say, an absolutist’s aversion to
out-groups but a contextualist’s rejection of a universalistic behav-
ioral code.17
To talk of a gene coded for the belief that universalistic behav-
ioral codes are improper (immoral?), is like talking about a
gene for one’s views concerning federalism. To propose a
gene coded for one’s belief regarding the proper balance
between states’ rights and the federal government defies com-
mon sense (and I will note the extent to which AFH appear
to have backed away from absurd claims of this sort in their
response to me, which I take as a positive development).
Perhaps it is engaging in philosophical reasoning of a
sort to point out that different kinds of explanation are
appropriate to different kinds of phenomena, and it is
only a misunderstanding of the phenomenon in question
that allows one to seek an inappropriate explanation for it.
䡬
The same is true in science. Quantum mechanics has very
little to tell us about the functioning of the human heart,
and if a physicist claimed that the resolution of remaining
difficulties with string theory promised greater understand-
ing of the etiology of heart disease, we would have to
conclude that he did not know what heart disease was.
(Note that this phenomenon, the “irreducibility” of our
scientific knowledge about the human heart to our scien-
tific knowledge about quantum mechanics, does not mean
that the heart is a mystical phenomenon, or lead to the
positing of a “heart-matter dualism” 18 ).
If I were asked, why did Napolean loose the Battle of
Waterloo, or why did the German Revolution of 1848
fail, or why did the practice of the racialized slavery of
blacks begin in Europe in the sixteenth century, or why
did democratic government first appear in ancient Ath-
ens, or why did the American Founders turn to ancient
Rome rather than Athens for their model of government,
and responded that advances in genetics held out the prom-
ise of an answer to these questions, the interlocutor would
have to conclude that I had no comprehension of what
history was. And note that this assertion does not turn
history into some kind of mystical phenomenon, or posit
a “mind-brain duality,” or any other such nonsense.
That those such as AFH do not adequately compre-
hend the phenomenon they are supposed to be investigat-
ing (political ideologies) is made abundantly clear by their 䡬
erroneous assumptions about the trans-historical and trans-
cultural nature of liberalism and conservatism as distinct
political ideologies, that liberalism and conservatism are
each accurately defined by the core cluster of attitudes
they list as comprising the liberal and conservative “phe-
notypes,” that the only noteworthy historical change in
these ideologies is what they are named, and that party
identification is a fixed variable at the present time in the
United States. It is only greedy reductionism, the ultimate
hubris, that impels political scientists to so egregiously
mischaracterize complex phenomena in order to fit them
into a reductionist explanatory model (and political sci-
ence has seen plenty of these in its checkered history).19 In
the words of the renowned geneticist, statistician, and evo-
lutionary biologist Richard Lewontin: “It is a sign of the
foolishness into which an unreflective reductionism can
lead us that we seriously argue from protein similarity to
political similarity.” 20
AFH and HH, along with many other political and social
scientists, suffer from massive confusion in failing to dis-
tinguish between the reasons why persons hold or believe
in specific political ideologies (i.e., the answer to the ques-
tion why does this individual hold the political ideology
she does) and political ideologies themselves. Whatever the
explanation as to why a given individual ultimately holds
the political ideology she does (and I will simply assert that
I believe the reasons to be potentially infinite, including,
e.g., upbringing, emotional appeal, rebellion against one’s
June 2008 | Vol. 6/No. 2 341
| |
Exchange | Politics, Genetics, and “Greedy Reductionism”
parents, attempts to please one’s parents, lazy and uncrit-
ical acceptance of the beliefs of those in her surrounding
environment, rigorous individual critical reflection and
rejection of the beliefs of those in her surrounding envi-
ronment, religious convictions, anti-religious convictions,
etc.), such an explanation does not tell us what political
ideologies are, how and why they developed at certain times
and places, how they were institutionalized in specific polit-
ical and social practices, and how they developed and trans-
formed over time. One might attempt to explain the
“etiology” of persons’ attitudes toward the U.S. Constitu-
tion on the basis of “genetic” personality traits, and con-
struct a “pro” and “anti” U.S. Constitution “phenotype.”
But a theory that purported to explain why any given
individual had the attitudes she did toward the U.S. Con-
stitution would not explain what the Constitution was, i.e.,
why it was written, when, and by whom, what political
principles it embodied or codified, and why.
Finally, if I engage in philosophy, so do AFH and HH,
even if they are not aware of it, or of the profound philo-
sophical assumptions concerning knowledge and reason that
underlie their beliefs regarding the nature of science on
the one hand, and political and moral beliefs on the other.
I cannot elaborate further on these points here, but will
simply end by posing a question to Hannagan and Hatemi.
If, as AFH claim, my political views are due largely to my
䡬 genes, for example, my views—perhaps mediated by genetic
personality traits—about Social Security, the proper scope
and limits of presidential power, and the war in Iraq, would
HH be willing to conjecture as to what percentage of their
views concerning genetics, e.g., the viability of twin stud-
ies, the genetic basis of political ideologies, the soundness
of the Equal Environment Assumption, are due to their
genes. Might they undertake a twin study to answer this
question? And if not, why?
Notes
The renowned geneticist Jonathan Beckwith, Ameri-
can Cancer Society Research Professor of Microbiol-
ogy and Molecular Genetics at Harvard Medical
School, has written a response to Alford, Funk, and
Hibbing and Hannagan and Hatemi which will
appear in a forthcoming issue of Perspectives on
Politics. His contribution is intended to supplement
my response, providing the unique perspective of a
geneticist in a debate that has been dominated by
political scientists and psychologists.
1 “Greedy reductionism” is a term coined by Daniel
Dennett (1995, 82) to condemn those forms of
reductionism that try to explain too much with too
little. Use of this expression is not meant to imply
an endorsement of Dennett’s thesis.
2 AFH 2008, 322.
3 Bouchard et al. 1990.
342 Perspectives on Politics
䡬
4 Charney 2008, 338.
5 Imagine if the validity of the theory of relativity
rested upon the accounts of Einstein’s mother as to
how she raised baby Einstein.
6 AFH 2008, 322.
7 Brody and Rothenberg 1988.
8 For both of these points, see, e.g., Carsey and
Layman 2006; Lockerbie 2002; Weisberg 1980,
2002; Brody and Rothenberg 1988; Mutz, Brody,
and Sniderman 1996; Norrander and Wilcox 1993;
Converse and Pierce 1992; Franklin 1992, 1984;
Niemi, Reed, and Weisberg 1991; Franklin and Jack-
son 1983; Fiorina 1981; Markus and Converse
1979.
9 Cooper and Zubek 1958.
10 AFH 2005, 164.
11 See, e.g., Bailey 1997.
12 AFH 2005, 161.
13 AFH 2005, 164. If it is not already obvious, the
impetus to mischaracterize political ideologies in this
manner comes from the desire to fit them wholly
within a reductionist genetic explanatory framework
that effectively bypasses history and culture. Just as,
e.g., the phenotype of hazel eyes, the result of a
corresponding genotype, is the same in all times and
all places (and can be characterized apart from any
specific cultural and historical context), so too,
according to AFH, with political ideologies.
14 AFH 2008, 325.
15 Ibid.
16 AFH 2005, 164.
17 Ibid., 165.
18 Neils Bohr was explicit that his denial of the reduci-
bility of biology to quantum mechanics had no
implications for “free will or determinism,” and did
not involve a mysticism incompatible with the true
spirit of science”; see Bohr 1936, 299.
19 Many kinds of “hubris” are exhibited in an article
such as AFH’s. Let me note just one other character-
istic “ethnocentric hubris”: In characterizing con-
temporary American liberalism and conservatism as
if they were the templates not only of liberalism and
conservatism in all of its varied historical and cul-
tural manifestations, but of all political ideologies in
all times and places, they equate the beliefs and be-
havior of contemporary Americans with the beliefs
and behavior of “humankind.” The genetic under-
pinning they give to such assumptions makes con-
temporary Americans the paradigm for humanity as
a biological species. As a helpful corrective to such
hubristic provincialism, I suggest that AFH travel to
the Amazon rain forest and undertake an extensive
study of the political attitudes of the remaining
indigenous tribes.
20 Lewontin 2001, 62.
䡬
| |
䡬
References
Alford, J.R., C.L. Funk, and J. Hibbing. 2005. Are
political orientations genetically transmitted? Ameri-
can Political Science Review 99 (2): 153–67.
_. 2008. Beyond liberals and conservatives to politi-
cal genotypes and phenotypes. Perspectives on Politics
6 (2): 321–28.
Bailey, R.C. 1997. Hereditarian Scientific Fallacies.
Genetica 99: 125–33.
Bohr, Neils. 1936. Causality and Complementarity.
Philosophy of Science 4. Address to Second Inter-
national Congress for the Unity of Science, June,
1936.
Bouchard, T.J. Jr., D. T. Lykken, M. McGue, N. L.Se-
gal, and A. Tellegen. 1990. Sources of human psycho-
logical differences: the Minnesota study of twins
raised apart. Science 250: 223–28.
Brody, Richard A., and Lawrence S. Rothenberg. 1988.
The instability of partisanship: An analysis of the
1980 presidential election. British Journal of Political
Science 18 (4): 445–65.
Carsey, Thomas M., and Geoffrey C. Layman. 2006.
Changing sides or changing minds? Party conversion,
issue conversion, and partisan change on the abortion
issue. American Journal of Political Science 50 (2):
464–77.
Charney, Evan. 2008. Genes and ideologies. Perspectives
on Politics 6 (2): 299–319.
Converse, Philip E., and Roy Pierce. 1992. Partisanship
and the party system. Political Behavior 14 (3):
239–59.
Cooper, R.M., and J.P. Zubek. 1958. Effects of enriched
and restricted early environments on the learning
ability of bright and dull rats. Canadian Journal of
Psychology 12: 159–64.
Dennett, Daniel. 1995. Darwin’s Dangerous Idea. Lon-
don: Penguin.
䡬
Fiorina, Morris P. 1981. Retrospective Voting in American
National Elections. New Haven: Yale University Press.
Franklin, Charles H. 1992. Measurement and the dy-
namics of party identification. Political Behavior 14
(3): 297–309.
_. 1984. Issue preferences, socialization, and the
evolution of party identification. American Journal of
Political Science 28 (3): 459–78.
Franklin, Charles H., and John E. Jackson. 1983. The
dynamics of party identification. American Political
Science Review 77 (4): 957–73.
Lewontin, Richard. 2001. It Ain’t Necessarily So: The
Dream of the Human Genome and Other Illusions.
New York: New York Review Books.
Lockerbie, Brad. 2002. Party identification: Constancy
and change. American Politics Research 30 (4):
384–405.
Markus, Gregory B., and Philip E. Converse. 1979. A
dynamic simultaneous equation model of electoral
choice. American Political Science Review 73 (4):
1055–70.
Mutz, Diana C., Richard A. Brody, and Paul M. Snider-
man, eds. 1996. Political Persuasion and Attitude
Change. Ann Arbor, MI: University of Michigan
Press.
Niemi, Richard G., David R. Reed, and Herbert F.
Weisberg. 1991. Partisan commitment: A research 䡬
note. Political Behavior 13 (3): 213–21.
Norrander, Barbara, and Clyde Wilcox. 1993. Rallying
around the flag and partisan change: The case of the
Persian Gulf War. Political Research Quarterly 46 (4):
759–70.
Weisberg, Herbert F. 1980. A multidimensional concep-
tualization of party identification. Political Behavior 2
(1): 33–60.
_. 2002. Partisanship and incumbency in presiden-
tial elections. Political Behavior 24 (4): 339–60.
June 2008 | Vol. 6/No. 2 343
 Twin Studies of Political Behavior: Untenable Assumptions?

Corey A. Morris
Department of Cell Biology
Harvard Medical School

&

Jon Beckwith
Department of Microbiology and Molecular Genetics
Harvard Medical School

Corey A. Morris, is a doctoral candidate in the Program of Biological and Biomedical

Sciences at Harvard Medical School (cmorris@hms.harvard.edu).

Jon Beckwith is American Cancer Society Professor in the Department of Microbiology

and Molecular Genetics, Harvard Medical School (jbeckwith@hms.harvard.edu).

Please direct correspondence to:

Jon Beckwith, Ph.D.

American Cancer Society Professor
Department of Microbiology and Molecular Genetics
Harvard Medical School
200 Longwood Avenue
Boston, Massachusetts 02115
jbeckwith@hms.harvard.edu
 John Alford, Carolyn Funk, and John Hibbing, in their American Political Science

Review articlei (hereafter referred to as “AFH”), employ twin studies to argue that

political ideologies (i.e., conservative or liberal political attitudes) are highly heritable

and thus strongly influenced by genetics. AFH state that genetic explanations of behavior

are “conspicuously absent” from the political science dialogue and charge that “political

scientists do not take seriously the possibility of nonenvironmental influences.”ii They

urge “political scientists to incorporate genetic influences”iii into models of political

behavior and to “join forces”iv with behavioral geneticists. Given the significant

criticisms of studies in which twins are used to understand the origins of complex human

behaviorsv and problems with the scientific data supposedly supporting the validity of

this method, such a move would be, at best, premature.

AFH argue that the classical twin approach for studying complex human social

behaviors is well supported by the evidence. Here, we examine the empirical studies

cited by AFH in support of a critical assumption underlying the validity of these twin

studies, the equal environment assumption (EEA).vi We do not claim that genetic factors

play no role in human behavioral traits, but rather we show that the empirical evidence

used in support of the underlying premise of these twin studies is weak, far less certain

than AFH would have readers believe. In fact, many of the studies cited as supporting

the validity of the twin method include data that violate predictions of the EEA. As a

result, the conclusions drawn on the basis of classical twin studies, as those presented by

AFH, are of dubious scientific value.

The Heritability Concept

1
 Heritability is a statistical term that should represent, for a trait of interest, the

fraction of variation in a population that is due to genetic contributions.vii Expressed as a

number ranging from 0.0 (no heritability) to 1.0viii (complete heritability), heritability is

measured for a specific population, at a specific time, interacting with a specific

environment.ix Thus, as pointed out by Schaffner, “if the environment changes, the

heritability will almost certainly change.”x This property of heritability is highlighted by

a study showing that IQ differences exhibit high heritability in families of higher

socioeconomic status, but near-zero heritability in families of lower socioeconomic

status.xi In other words, in an advantaged population, a large fraction of IQ variability

was attributed to genetic influences, while in a disadvantaged population, IQ variability

was attributed almost entirely to environmental influences. In effect, incorporating

measures of environmental influences into such studies can yield dramatic variations in

heritability estimates. Relatively few such twin studies have treated environment very

deeply,xii usually attempting to rule it out as having a confounding effect on genetic

conclusions. As leading behavioral geneticist Eric Turkheimer has put it, this is “hardly

the stuff of good environmental analysis.”xiii

The “Crucial” Equal Environments Assumption

AFH acknowledge that the equal environment assumption is “crucial to

everything that follows from twin research.”xiv They conclude that the “caveats” raised

have “been subject to sustained and varied investigation and…[have] been found to hold

up under empirical scrutiny.”xv In support of this claim, they cite studies that look at the

effects of measures of similarity of appearance, treatment, and environment (including

2
so-called “misperceived zygosity studies”) on shared traits between twins.xvi We review

the evidence that they argue supports their conclusions.

Studies in which heritability of behavioral traits is based on comparisons of pairs

of monozygotic (MZ) and dizygotic (DZ)xvii twins rely on the validity of the EEA. The

EEA assumes that MZ and DZ twins each share environments to the same extent. In

particular, it is assumed that the two types of twins do not differ in those environmental

influences that might affect the trait under study (trait-relevant environments). However,

critics argue that, because of their identical appearance, MZ twins are treated and

interacted with by parents and the outside world much more similarly than are DZ twins

or ordinary siblings. Consider the potential influence of shared physical features of MZ

twins on social interactions. Children who are obese, for example, might suffer

indignities that others do not. Children thought of as “attractive” might experience a very

different world than those deemed “ugly.” MZ twins on average share closer bonds than

DZ twins by a number of measures. Both this suggested closer bond between MZ twins

and the potentially similar responses of family and society to the identically-featured MZ

twins could well influence the behavioral development of the children in the direction of

greater similarity. If this were to be the case, the incorrect acceptance of the EEA could

lead to heritability measurements in twin studies that would be highly inflated, despite the

fact that the trait under study was strongly environmentally influenced.

Thus, in order for heritability estimates to be valid for a particular trait, MZ and

DZ twins should experience equal trait-relevant environments. This is a primary problem

in the studies cited by AFH, since determining which environmental factors to assess may

be difficult to achieve. After all, such studies are meant to sort out the factors that

3
influence a trait, and should not assume to know a priori all of the relevant environmental

influences.xviii Even those researchers who do attempt to measure environmental

influences, as the authors of one study point out, choose out of “a virtually endless array

of possible environmental characteristics.”xix Indeed, none of the studies we reviewed

here sought specific “trait-relevant” environmental effectors, but only attempted to rule

out effects of general measures of environmental similarity.

“Misperceived Zygosity” Studiesxx

Misperceived zygosity studies, considered to be tests of the EEA, take advantage

of twin pairs for which their twin status (i.e., MZ or DZ) has been wrongly assigned.

These studies are made possible by instances where biologically identical twins call

themselves or are called by their parents “fraternal,” or where fraternal twins call

themselves or are called by their parents “identical.” Trait correlations for these

misidentified twins are then compared to those for correctly identified twins in order to

test the effects of “true zygosity” (presumed to measure a genetic effect) versus

“perceived zygosity” (presumed to measure an environmental effect). Thus, the

argument goes, if MZ twins who perceive themselves to be DZ are just as similar in trait

measures as MZ twins who correctly identify themselves (or DZ twins who misidentify

as MZ just as dissimilar as DZ twins who correctly identify themselves), then the

environmental influence of more similar treatment for MZ twins can be concluded “to be

at best irrelevant.”xxi

However, in order for “perceived zygosity” to be a meaningful test of the EEA, it

must be a true surrogate for the environmental similarity a pair of twins’ experiences.

4
For example, a genetically identical twin pair who misidentify themselves as “fraternal”

must believe themselves as well as be perceived by others to be non-identical fraternal

twins. Otherwise, this may merely be a case of genetically MZ twins, who are likely to

appear remarkably similar, but confused or were incorrectly told their twin status. Unless

self-identification of zygosity is shown to correlate with beliefs about degree of similarity

and actual physical similarity, it is questionable whether merely mis-categorization is a

meaningful measure of how a twin pair will be treated, when MZ twins will still appear

physically similar and DZ twins dissimilar. Nevertheless, the authors of these studies,

without evidence, assume that DZ twin pairs misidentified as MZ pairs “view themselves

as more similar [and] will be treated more alike by parents, family, and society.”xxii This

is an untested assumption. We propose the term “misidentified zygosity” as more

appropriate for such twin pairs, as “misperception” has not been demonstrated.

In fact, data in two “misperceived zygosity” studies appear to contradict the

notion that physical similarity is the basis for the perception of zygosity status. One

study found that, compared to MZ or DZ twins who were correct about their zygosity,

MZ twins who incorrectly believed they were DZ “were much less likely to base their

opinion on physical appearance, but more likely to base their opinion on what their

parents were told at their birth.”xxiii Another study found that “when mothers were

mistaken about the zygosity of their twins, they usually relied on information given to

them from a doctor at the time of the twins’ birth,” as opposed to similarity of physical

appearance.xxiv These findings raise questions about the utility of using “misperceived

zygosity” as a test of the EEA. As the authors of one study using this method

5
acknowledge, “a single measure of self-perceived zygosity may not be the most accurate

representation of what critics mean by biasing perception of the two kinds of twins.”xxv

The limitation of using “perceived zygosity” as a surrogate for environmental

similarity is revealed by the data presented in a 2002 study.xxvi MZ twins sharing classes

were significantly more likely to share symptoms of separation anxiety disorder, attention

deficit-hyperactivity disorder (ADHD), and oppositional defiant disorder. However,

“perceived zygosity” was not associated with shared symptomology, indicating that

misidentified zygosity did not capture environmental effects that were seen when

accounting for the likelihood of sharing classes. Furthermore, sharing friends by DZ

twins predicted similar ADHD scores, although a “perceived zygosity” score for shared

ADHD did not have a statistically significant effect. This suggests that “perceived

zygosity,” as measured by these studies, is unlikely to be a good indicator of

environmental similarity, and thus will fail to truly test the EEA.

Violations of the EEA

One of the clearest and most consistent results of the studies cited by AFH was

that MZ twins are indeed more similar on measures of environmental and physical

similarity. MZ twins were “markedly more confusable in appearance than DZ twins;”xxvii

significantly more likely than DZ twins to report being treated “as two of a kind” by

parents, to have dressed identically, to have been in the same class at school, to have

played together “almost always,” and to have spent time together “almost always”;xxviii

and share a much more similar environment based on mothers’ accounts of childhood

environmental similarity (including sharing friends, sharing classes, and dressing

6
alike).xxix As the authors of one study state, “we must conclude that the MZ twins in our

sample were treated more similarly than were the DZ twins and that therefore the equal

environments assumption was violated.”xxx

Ultimately, the authors of all of these studies conclude that despite apparent

violations of the EEA generally, their measures of environmental similarity did not

influence the traits under study, and therefore “trait-relevant” EEA was not violated.

However, a closer look at the studies in support of the EEA finds that many actually

contain significant data arguing that “trait-relevant” EEA was indeed violated. For

example, one study found that DZ twins who were more likely to be confused in

appearance were more likely to have similar scores for activity and impulsivity traits.

MZ and DZ twins as a whole who were more similar in appearance were more likely to

share all four measures of personality traits measured.xxxi

In another study, the authors report that “for personality measures MZ twins were

significantly more similar than DZs, and DZ twins who believed they were monozygotic

were more similar than those…who correctly believed they were fraternal pairs.” In

other words, “both true and perceived zygosity were related to cotwin similarity on

personality measures.”xxxii Furthermore, they found that physical similarity (as assessed

by eight raters) predicted more similar Raven IQ scores for monozygotic twins. The

authors admit their data is evidence of “bias in perceived and physical similarity creating

greater cognitive similarity among MZ twins,” although they downplay the effect.

Nevertheless, they do conclude, “for personality variables, perceived zygosity may have

some effect on fraternal pairs who believe themselves to be monozygotic.”xxxiii

7
 Another of these studies found significant associations between measures of DZ

twins’ preferences for similar treatment, similarity of environment, and preference for

similar experience with anxiety and depression scores.xxxiv For MZ twins, an association

was found between measures of environment and depression. The authors conclude that

the “equal environments assumption appears to be invalid.”xxxv Likewise, a different

study found that measures of environmental similarity predicted greater similarity in

symptom scores for nearly half of the tested traits.xxxvi For example, MZ twin pairs who

were more frequently in the same classes were significantly more likely to share similar

symptom scores for all of the measured disorders than those who were in the same

classes less often. Dressing alike predicted greater symptom correlation among MZ twins

for separation anxiety disorder, ADHD, or conduct disorder.

Further indicating that environmental similarity of MZ twins confounds twin

studies is the fact that the data also show that DZ twins who more often share friends,

dress alike, or share classes, were significantly more similar on scores for ADHD,

separation anxiety disorder, and conduct disorder, respectively.xxxvii In addition, they

found a significant effect of “perceived zygosity” for MZ twins on shared conduct

disorder, and for DZ twins on shared separation anxiety and conduct disorders. These

statistically significant findings call into question the EEA. This data is all but ignored

by the authors of this study, and where the effect of environmental similarity is discussed,

it is downplayed.xxxviii

If environmental similarity does influence MZ twin concordance for a trait in

question, then this would be expected to inflate estimates of heritability. Accounting for

the unequal environments would decrease estimated heritability. In a study measuring

8
the “contact frequency” of twin pairs, the authors found that heritability estimates

dropped dramatically when the samples were stratified by level of contact between

twins.xxxix On average, their heritability measures for twins with infrequent contact

dropped 31 percent below twins with frequent contact. Compared with the total sample

of twins, controlling for the frequency of contact indicates that their measures of

heritability were over-estimated on average by 27 percent. The authors note that “the

heritability estimates were consistently higher among twin pairs with frequent contact,

suggesting a potential violation of EEA.”xl

Given that several of these studies report findings that appear to violate both

general and trait-relevant EEA, how do all of these studies conclude that the EEA is

valid?xli Despite small sample sizes and weak statistical power, several studies do so

solely on the basis of negative results, or a failure to find statistically significant

correlations between their measures of similarity and the trait in question.xlii Not all of

the studies, however, based their EEA-validating conclusions on negative results. One

study presenting significant evidence that the EEA did inflate heritability estimates

nevertheless emphatically concluded that they “found no support for violation of

EEA.”xliii In order to draw this conclusion, they used 40,000 simulated twin pairs and the

new assumption that MZ twins select their environment on the basis of their genetics.xliv

As if to highlight their conclusions in spite of their actual empirical data, they state that

“the simulation analyses thus provide an interpretation of our empirical results without

violation of the EEA.”xlv

Retrospective Nature of the Studies

9
 Another problem with studies evaluating the validity of the EEA is that important

data is often obtained from retrospective information. One study asked adult twins with a

mean age of 34 how they were treated as children. As the authors put it, the subjects

were “reporting on events and their feelings about those events some 20 years after their

occurrence.”xlvi Another study measured emotional and behavioral problems via mother-

reported retrospective telephone interviews, recording mothers’ reports of twins’

zygosity, as well as recounts of childhood environmental similarity.xlvii This

retrospective format is suspect, given that even in a study where parents gave

contemporary reports of their children, mothers and fathers independently agreed on the

answers only about one-third of the time. When asked the same questions only three

months later, their answers changed over 25 percent of the time.xlviii This calls into

question the reliability of retrospective data over a matter of weeks, let alone decades.

In addition to methodological flaws, all of the studies cited in support of the EEA

used small samples, ranging in size from 41 twin pairsxlix to 201 twin pairsl having

“misperceived zygosity.” One of the largest studies (with 3155 total twin pairs, 185 of

which had “misperceived zygosity”) included a power analysis indicating that even that

sample size offered weak to no statistical power to detect an influence of misidentified

zygosity.li A study cited by AFH summarizes the statistical weakness of these earlier

reports, noting that studies supporting the EEA may have reached their conclusions

“because of limited statistical power” and “because the validity of the assumption is

usually established by failing to find a significant effect.” lii

Conclusion

10
 We agree with the authors AFH when they state that the EEA is essential for

conclusions drawn from twin studies concerning human behavioral traits. However, we

disagree that the EEA has been tested and validated. We point out numerous problems

with the studies cited by AFH. These studies report small sample sizes with low

statistical power, problematic retrospective interviews, and assumptions about

misidentified zygosity that are often not supported by data. The attempts to control for

trait-relevant environments in these efforts are quite limited, admittedly because it is

difficult to know exactly which environmental factors to control for. Furthermore,

several of the studies cited in support of the EEA contain evidence that could be seen as

arguing against the assumption. It is difficult not to get the sense that in the face of

inconvenient data, a fair amount of intellectual acrobatics was necessary to arrive at some

of the conclusions drawn.

AFH have made a plea for the political science research community to take more

seriously twin studies that suggest a strong genetic component affecting political beliefs.

While, to many, twin studies seem to present an ideal opportunity to study the relative

contributions of genetics and environment to human behavioral traits, these studies face

many more problems than has been admitted. In a research area that is often

misrepresented to the public, we urge political scientists to take a more critical look at the

studies that supposedly provide the foundation for this field.

Finally, we point out that in recent years, some researchers in the field of human

behavioral genetics have begun to take more seriously the role of environment in

influencing human behavioral traits. We have already referred to the studies of Eric

Turkheimer and colleagues. Other examples include studies that have incorporated

11
measurements of child abuse into a study of genetic effects on anti-social behavior.liii A

study genuinely accounting for social factors found that certain measures of greater

similarity among MZ twins compared to DZ twins diminished or disappeared.liv The

findings of James Flynn and his colleagueslv that the mean IQ of populations in many

countries (including the U.S.) has been steadily increasing over the last 50 years raise

questions about the more deterministic views of genetic influences on human cognition.

A recent article looking at concentrated disadvantage found neighborhood-level effects

on verbal ability among African-American childrenlvi and indicates more sophisticated

measures of environment may be necessary to truly detect significant environmental

influences on complex human traits.

i
Alford et al. 2005.
ii
Ibid.
iii
Ibid.
iv
Alford et al. 2008.
v
i.e., Pam et al. 1996, Joseph 1998, Kamin and Goldberger 2002, Rutter 2002, Horwitz et

al. 2003a and 2003b, Beckwith 2006, and Ehrlich and Feldman 2007.
vi
The primary studies reviewed include Plomin et al. 1976, Scarr and Carter-Saltzman

1979, Morris-Yates et al. 1990, Kendler et al. 1993, Xian et al. 2000, Cronk et al. 2002,

and Eriksson et al. 2006.

vii
For an excellent discussion of heritability, see Sober 2001, 47-78.
viii
This estimated statistic gives the false sense that “heritability” is concrete and fixed.

12
ix
It is important to keep in mind that twin studies in principle allow one to make claims

regarding the effects of both genes and environment without ever having actually

measured anything about genes or the environment.

x
Schaffner 2006, 16.
xi
Turkheimer et al. 2003.
xii
Three notable exceptions include 1) a study looking at the environments of twins’ first

year of life, finding that twins disproportionately experience a number of developmental

circumstances that raise questions about generalizing twins studies to the population at

large (Ainslie et al. 1987); 2) a study indicating that intrauterine environment may bias

psychological and behavioral estimates of heritability (Prescott et al. 1999); and 3) a

study demonstrating social sources of shared twin behavior previously attributed largely

to genetic influences (Horwitz et al. 2003a and 2003b).

xiii
Turkheimer 2006, 102.
xiv
Alford et al. 2005.
xv
Ibid.
xvi
AFH also cite Bouchard et al. (1990), a study purporting to test the notion that if

common environment is of significance, then “twins reared apart” should be less alike

that twins reared together. We have addressed this study elsewhere (Beckwith et al.

1991), finding twins’ reared apart status to be highly questionable. Also see Richardson

1998, 140-145 and Joseph 2004, 97-136 for critical review of Bouchard et al. 1990 and

the history of “twins reared apart.”

xvii
The terms “MZ” and “DZ” are used throughout this commentary, except when

discussing studies that refer explicitly to “identical” and “fraternal” twins.

13
xviii
Beckwith 2006, 79.
xix
Cronk et al. 2002.
xx
In the interest of space we have not been able to include a full accounting of the

methodological limitations of these studies, and have focused on what we believe to be

some of the more important points. Other issues that arise include biased sampling,

insufficient variation for analyses of variance, and crude measures of “contact

frequency.”
xxi
Alford et al. 2005.
xxii
Xian et al. 2000.
xxiii
Kendler et al. 1993.
xxiv
Cronk et al. 2002.
xxv
Scarr and Carter-Saltzman 1979.
xxvi
Cronk et al. 2002.
xxvii
Plomin et al. 1796.
xxviii
Morris-Yates et al. 1990.
xxix
Cronk et al. 2002.
xxx
Morris-Yates 1990. Although their end conclusion is that the EEA is valid.
xxxi
Plomin et al. 1976.
xxxii
Scarr and Carter-Saltzman 1979.
xxxiii
Ibid.
xxxiv
Morris-Yates et al. 1990.
xxxv
Ibid.
xxxvi
Cronk et al. 2002.

14
xxxvii
Ibid.
xxxviii
Ibid.
xxxix
Eriksson et al. 2006. Measuring the heritability of ”physical activity.”
xl
Ibid.
xli
The authors of these studies, for example, conclude that violations of EEA “do not

appear to bias twin studies in the direction of inflated heritabilities” (Plomin et al. 1976);

that “the critical assumption of equal environmental variance for MZ and DZ twins is

tenable” (Scarr and Carter-Saltzman 1979); and that “similar treatment imposed upon MZ

twins on the basis of their zygosity alone is…not a threat to the validity of the twin

method” (Morris-Yates et al. 1990).

xlii
Plomin et al. 1976, Scarr and Carter-Saltzman 1979, Morris-Yates et al. 1990, Kendler

et al. 1993, and Cronk et al. 2002.

xliii
Eriksson et al. 2006.
xliv
This has been a common appeal made by twin study researchers when faced with the

issue of environmental “confounders” of genetic interpretations—what we call the

“Geneticization of Environment.”
xlv
Eriksson et al. 2006.
xlvi
Morris-Yates et al. 1990.
xlvii
Cronk et al. 2002.
xlviii
Plomin et al. 1976.
xlix
Scarr and Carter-Saltzman 1979.
l
Cronk et al. 2002.
li
Xian et al. 2000.

15
lii
Cronk et al. 2002.
liii
Although the replicability of these studies have been mixed (Morris et al. 2007).
liv
Horwitz et al., 2003a and 2003b.
lv
Flynn 2007.
lvi
Sampson et al. 2008.

16
 REFERENCES

Ainslie, Ricardo C., Kelly M. Olmstead, and Daniel D. O’Loughlin. 1987. “The early

developmental context of twinship: some limitations of the equal environments

hypothesis” American Journal of Orthopsychiatry 57 (1): 120-4.

Alford, John R., Carolyn L. Funk, and John R. Hibbing. 2005. “Are political orientations

genetically transmitted?” American Political Science Review 99 (2): 153-68.

Alford, John R., Carolyn L. Funk, and John R. Hibbing. 2008. “Beyond Liberals and

Conservatives to political genotypes and phenotypes” Perspectives on Politics 39 (1): this

issue.

Beckwith, Jon, Lisa Geller, and Sahotra Sarkar. 1991. “IQ and Heredity” [Letter to the

Editor] Science 252 (5003): 191.

Beckwith, Jon. 2006. “Wither human behavioral genetics?” In Wrestling with Behavioral

Genetics: Science, Ethics, and Public Conversation, ed. E. Parens, A. Chapman, and N.

Press. Baltimore: The Johns Hopkins University Press.

Bouchard, Thomas J., Jr., David T. Lykken, Matthew McGue, Nancy L. Segal, and Auke

Tellegen. 1990. “Sources of human psychological differences: The Minnesota Study of

Twins Reared Apart” Science 250 (4978): 223-8.

17
Cronk, Nikole J., Wendy S. Slutske, Pamela A.F. Madden, Kathleen K. Bucholz, Wendy

Reich, and Andrew Heath. 2002. “Emotional and behavioral probelems among female

twins: An evaluation of the equal environments assumption” Journal of the American

Academy of Child and Adolescent Psychology 41 (7): 829-37.

Ehrlich, Paul, and Marcus W. Feldman. 2007. “The fallacy of genetic reductionism”

Dædalus Spring: 5-12.

Eriksson, Marit, Finn Rasmussen, and Per Tynelius. 2006. “Genetic factors in physical

activity and the equal environment assumption—the Swedish Young Male Twin Study”

Behavior Genetics 36 (2): 238-47.

Flynn, James R. 2007 What is Intelligence? Cambridge: Cambridge University Press.

Horwitz, Allan V., Tami M. Videon, Mark F. Schmitz, and Diane Davis. 2003a.

“Rethinking twins and environments: Possible sources for assumed genetic influences in

twin research” Journal of Health and Social Behavior 44 (2): 111-29.

Horwitz, Allan V., Tami M. Videon, Mark F. Schmitz, and Diane Davis. 2003b. “Double

vision: Reply to Freese and Powell” Journal of Health and Social Behavior 44 (2): 136-

41.

18
Joseph, Jay. 1998. “The Equal Environment Assumption of the Classical Twin Method:

A Critical Analysis” Journal of Mind and Behavior 19 (3): 325-58.

Joseph, Jay. 2004. The Gene Illusion. New York: Algora Publishing.

Kamin, Leon J., and Arthur S. Goldberger. 2002. “Twin studies in behavioral research: A

skeptical view” Theoretical Population Biology 61 (1): 83-95.

Kendler, Kenneth S., Michael C. Neale, Ronald C. Kessler, Andrew C. Heath, and

Lindon J. Eaves. 1993. “A test of the equal-environment assumption in twin studies of

psychiatric illness” Behavior Genetics 23 (1): 21-7.

Morris, Corey, Aimee Shen, Khadija Pierce, and Jon Beckwith. 2007. “Deconstructing

violence” GeneWatch 20 (2): 3-9.

Morris-Yates, Allen, Gavin Andrews, P. Howie, and S. Henderson. 1990. “Twins: a test

of the equal environments assumption” Acta Psychiatrica Scandinavica 81 (4): 322-6.

Pam, Alvin, Susan S. Kemker, Colin A. Ross, and R. Golden. 1996. “The ‘Equal

Environments Assumption’ in MZ-DZ twin comparisons: An untenable premise of

psychiatric genetics?” Acta geneticae medicae et gemellologiae 45 (3): 349-60.

19
Plomin, Robert, Lee Willerman, and John C. Loehlin. 1976. “Resemblance in appearance

and the equal environments assumption in twin studies of personality traits” Behavior

Genetics 6 (1): 43-52.

Prescott, Carol A., Ronald C. Johnson, and John J. McArdle. 1999. “Chorion type as a

possible influence on the results and interpretation of twin study data” Twin Research 2

(4): 244-9.

Richardson, Ken. 1998. The Origins of Human Potential. London: Routledge.

Rutter, Michael. 2002. “Nature, nurture, and development: From evangelism through

science towards policy and practice” Child Development 73 (1): 1-21.

Sampson, Robert J., Patrick Sharkey, and Stephen W. Raudenbush. 2008. “Durable

effects of concentrated disadvantage on verbal ability among African-American children”

Proceedings of the National Academies of Sciences 105 (3): 845-52.

Scarr, Sandra and Louise Carter-Saltzman. 1979. “Twin method: defense of a critical

assumption” Behavior Genetics 9 (6): 527-42.

Schaffner, Kenneth F. 2006. “Behavior: Its nature and nurture” In Wrestling with

Behavioral Genetics: Science, Ethics, and Public Conversation, ed. E. Parens, A.

Chapman, and N. Press. Baltimore: The Johns Hopkins University Press.

20
Sober, Elliott. 2001. “Separating nature and nurture” In Genetics and Criminal Behavior,

ed. D.T. Wasserman and R.S. Wachbroit. Cambridge: Cambridge University Press.

Turkheimer, Eric, Andreana Haley, Mary Waldron, Brian D’Onofrio, and Irving I.

Gottesman. 2003. “Socioeconomic status modifies heritability of IQ in young children”

Psychological Science 14 (6): 623-8.

Turkheimer, Eric. 2006. “Mobiles: A gloomy view of research into complex human

traits” In Wrestling with Behavioral Genetics: Science, Ethics, and Public Conversation,

ed. E. Parens, A. Chapman, and N. Press. Baltimore: The Johns Hopkins University

Press.

Xian, Hong, Jeffrey F. Scherrer, Seth A. Eisen, William R. True, Andrew C. Heath, Jack

Goldberg, Michael J. Lyons, and Ming T. Tsuang. 2000. “Self-reported zygosity and the

equal-environments assumption for psychiatric disorders in the Vietnam Era Twin

Registry” Behavior Genetics 30 (4): 303-10.

21