Abstract: Social scientists most often seek to empirically validate something already observed. Genetics identifies the unobserved. It provides a starting point to identify developmental pathways to preferences and behaviors. Understanding differences in the genome can help identify why people who experience the same social environment physically perceive it differently and react to it differently. The introduction of evolutionary theory, combined with methods and approaches from genetics, genomics, epigenetics, and molecular biology, has substantially changed the way in which social scientists explore and understand the development and maintenance of political values and behaviors. This chapter reviews findings from recent empirical and theoretical studies that have explored how genetic factors account for some part of why people differ politically.
Keywords: evolution, genetics, politics, attitudes, nature–nurture
---
Nature, Nurture, and Politics
Since
antiquity, philosophical debates have persisted as to whether it is
nature or nurture that guides humanity. Today, we know, it is neither
alone. Empirical findings across the natural and social sciences have
provided compelling evidence that simplistic monocausal biological or
environmental explanations for complex social traits are implausible.
Instead, in order to have any true understanding of the origins and
development of complex behaviors, we must focus on developmental
processes that treat nature and nurture as interdependent (Lewkowicz, 2011).
Universal human traits and
variance in such traits, including political values, are the product of
the complex interaction between inherited physiology, socialization, and
personal experiences. The process is not linear but, rather, recursive.
Individuals’ inherited biology and experiences influence how they
interconnect with their social world, which in turn activates,
represses, and conditions psychological and biological responses (McDermott & Hatemi, 2013; Ridley, 2003).
These then, of course, condition, alter, and change one’s experiences,
including the environments people select into and the way people see,
perceive, and interpret their social world, and thus how they react to
it and how others react to them—creating an indefinite circle of
development.
Not all are on board with
this position. Indeed, remarkably, after thousands of years and
overwhelming evidence to the contrary, the nature versus nurture
dichotomy continues to doggedly persist in modern academic and public
discourse (Eagly & Wood, 2013).
Although it is difficult to find a scientist who believes any complex
social trait or behavior is all nature, there are still social
scientists who argue that it is all nurture (Charney, 2008a; Shultziner, 2013).
Nevertheless, when it comes to the majority of the social sciences, the
tide has turned. During the past 40 years, research has eroded
(p. 282)
the belief in such a dichotomy, and during the past 10
years in particular, this erosion appears to be leading to a complete
collapse of the division (Hatemi & McDermott, 2012a).
Today, most social scientists
are no longer asking whether genes influence political behavior.
Instead, the discussion has moved on to asking why, what, when, and to
what extent particular environments trigger or repress specific genetic
processes (Benjamin et al., 2012; Hatemi, Byrne, & McDermott, 2012; Hatemi & McDermott, 2012a).
A surging wave of energy has been directed at elucidating the influence
of genetic factors on political attitudes and behaviors, including vote
choice (Dawes & Fowler, 2009; Fowler, Baker, & Dawes, 2008; Fowler & Dawes, 2008; Hatemi, Medland, Morley, Heath, & Martin, 2007; Littvay, Weith, & Dawes, 2011), political ideologies (Alford, Funk, & Hibbing, 2005; Hatemi, Eaves, & McDermott, 2012), political attitudes (Alford et al., 2005; Cranmer & Dawes, 2012; Hatemi, Funk, et al., 2009; Hatemi & McDermott, 2016; Hatemi et al., 2010, 2014; Smith & Hatemi, 2013), party identification and identity (Dawes & Fowler, 2009; Fazekas & Littvay, 2012; Hatemi, Alford, Hibbing, Martin, & Eaves, 2009; Settle, Dawes, & Fowler, 2009; Weber, Johnson, & Arceneaux, 2011), political trust (Merolla, Burnett, Pyle, Ahmadi, & Zak, 2013; Ojeda, 2016; Oskarsson, Dawes, Johannesson, & Magnusson, 2012; Sturgis et al., 2010), political participation (Dawes et al., 2014; Fazekas & Hatemi, 2016; Fowler et al., 2008; Fowler, Dawes, & Settle, 2011; Klemmensen, Hatemi, Hobolt, Petersen, et al., 2012), sophistication and efficacy (Arceneaux, Johnson, & Maes, 2012; Klemmensen, Hatemi, Hobolt, Petersen, et al., 2012; Klemmensen, Hatemi, Hobolt, Skytthe, et al., 2012), and aggression and conflict (McDermott, Dawes, Prom-Wormley, Eaves, & Hatemi, 2013; McDermott & Hatemi, 2017; McDermott, Tingley, Cowden, Frazzetto, & Johnson, 2009; Stam, Von Hagen-Jamar, & Worthington, 2012).
A number of reviews on evolutionary theory, genetic theory, and what
genes are and how they operate, in addition to primers on advanced
statistical and molecular genetics, are present in the social science
literature (Barkow, Cosmides, & Tooby, 1992; Fowler & Schreiber, 2008; Hatemi, Byrne, et al., 2012; Hatemi & McDermott, 2011a, 2011b; Hibbing & Smith, 2007; Lopez, McDermott, & Petersen, 2011; Lumsden & Wilson, 1981; McDermott & Hatemi, 2013; Medland & Hatemi, 2009; Rushton, Littlefield, & Lumsden, 1986; Smith, Larimer, Littvay, & Hibbing, 2007; Verhulst & Hatemi, 2013).
Also, funded workshops on genetic analyses are available annually
through multiple mediums specifically tailored for social scientists.
The creation of the link
between genetic inherence and social traits promises to be one of the
most exciting and productive avenues of social science inquiry in the
21st century. It has not gone unnoticed that over the course of the 20th
century, the natural sciences greatly outperformed the social sciences.
King (2011) noted,
Fifteen years ago, Science published predictions from each of 60 scientists about the future of their fields. The physical and natural scientists wrote about a succession of breathtaking discoveries to be made, inventions to be constructed, problems to be solved, and policies and engineering changes that might become possible. In sharp (p. 283) contrast, the (smaller number of) social scientists did not mention a single problem they thought might be addressed, much less solved, or any inventions or discoveries on the horizon. Instead, they wrote about social science scholarship, how we once studied this and in the future we’re going to be studying that. (p. 719)
Between 1900 and 2000,
aerospace went from having never successfully flown an airplane to
putting a man on the moon. Medical science extended the average human
life expectancy by decades. Engineering went from steam engines to
bullet trains, the Internet, and artificial intelligence. Even
meteorologists, formerly the subject of jokes, now reliably and
consistently forecast the weather. In contrast, the social sciences have
made relatively little progress. Economists have been studying the
causes of the Great Depression for almost 100 years but could not
predict the 2007–2008 global financial crisis. Experts in international
relations have been studying the causes of World War I but were totally
blindsided by the fall of the Berlin Wall and modern Islamic terrorism.
Others have been studiously researching American voting behavior for 100
years, but before late spring 2016, few believed, much less predicted,
that Donald Trump would win the Republican nomination and the
presidency.
We are, of course, being too
hard on our discipline. The natural sciences have many advantages in
comparison. One that we might be able to leverage, however, is a
universal theory of behavior—evolution. In the natural sciences,
contradictory theories tend not to survive long, even between different
branches (e.g., chemistry and geology). They tend to be lightning rods
for empirical researchers who rush to resolve the debate. The speed of
publication, the focus on evidence, the willingness to let go of
established wisdom, and the focus on innovation and discovery comprise
the nature of the natural sciences. The same has not been true of the
social sciences. Keynesian and Hayekian economic theories exist side by
side, like realism and liberalism continue to coexist in international
relations, and rational choice and structuralism in political science.
The contending schools of thought in the social sciences more often
speak past one another rather than to each other, which retards the
advancement of disciplinary knowledge and often impedes
interdisciplinary research.
Genetically informed social
science research has the potential to cut through many existing
cleavages. It may provide social scientists with a common foundation on
which explanations of social behavior can be built. Empirical research
in genetics demands replication, and it is in a constant and dynamic
state of advancement, with new methods and approaches developed almost
daily. Inclusion of genetics requires social scientists to move beyond
all-encompassing and often conflicting paradigmatic worldviews and begin
to have the same conversations the natural sciences were having
throughout their hyperproductive 20th century. Already, genetically
informed research is having an impact on many areas of political science
research, including providing rational choice theories with some hint
of where preferences originate, which has long puzzled the field (Benjamin et al., 2012; Camerer, Loewenstein, & Prelec, 2005; Cesarini, Dawes, Johannesson, Lichtenstein, & Wallace, 2009; Dawes et al., 2012; Hatemi & McDermott, 2011b; Hibbing, Smith, & Alford, 2013).
(p. 284)
Considering the importance and promise of genetic
approaches to the social sciences, this chapter offers a review for
those new to the area. In so doing, an extensive list of references is
provided to assist readers in delving deeper into the area.
The Why, What, and How of Genetics and Politics
Genetics offers a suite of
methods to better explicate the pathways that lead to variation on
traits of interest. That is, genetics provides tools to better
understand why individuals differ and why people are also the same. They
do not replace other methods but provide additional information. The
rapidly expanding research on genetics and political science can be
categorized a number of different ways. The most popular have been
either by research method (e.g., classical twin studies vs. genome
studies; see Hatemi, Dawes, Frost-Keller, Settle, & Verhulst, 2011) or by traits (e.g., political attitudes vs. vote choice; see Hatemi & McDermott, 2012a).
In this chapter, we take a more functional approach by cataloging the
recent literature on the topic and dividing it along lines of its
overall “research objectives.” Simply stated, we categorize the
literature by studies that ask “Why?” “What?” or “How?”
A first avenue of research
has asked “why” the human genome has evolved to exercise an influence on
political traits. This research avenue has primarily been theoretical
and deductive and has employed evolutionary reasoning and primate
observations to deduce the origins of genetic influences on political
attitudes (Lockyer & Hatemi, 2014; Lopez et al., 2011; Petersen, 2010; Proctor & Brosnan, 2011).
Generally, this research agenda assumes that modern humans have evolved
to display political traits; because these traits have been “selected
in,” then they must have served a purpose through human evolutionary
history. This assumes that modern political interactions are reflections
of tribal living. Hence, deductively reasoning as to “why” humans have
evolved to have genetically influenced political traits may point
empirical researchers toward why human systems are similar across
diverse ecologies.
The second research grouping
has attempted to determine “what” political traits (e.g., voting) are
genetically influenced. This research is akin to identifying omitted
independent variables on discrete political behaviors and more than the
other research agenda has been responsible for showing that significant
biological influences exist on political traits. Not only has it been
extremely successful in identifying that genetic influences operate on a
range of different political beliefs and behaviors but also, perhaps
more important, this research agenda has encouraged the wider
exploration of the effects of neurobiology, hormones, and other
physiological characteristics on political traits (Hatemi &
McDermott, 2011b, 2012a; Hibbing et al., 2013; McDermott & Hatemi, 2013; Mooney, 2012; Shenkman, 2016; Tuschman, 2013; Weeden & Kurzban, 2014).
Finally, there is a nascent
stream of research that is inquiring into “how” and which genes
influence specific political traits. This research is attempting to
reveal the
(p. 285)
neurological and chemical pathways that link the genes
and regions of the genome with the eventual political trait. Researchers
have used advanced imaging technology and samples of people with brain
damage or abnormalities to attempt to trace neurological and
neurochemical pathways from DNA to the expressed political trait. This
research is in its early stages and holds much promise of bringing true
discovery back into the social sciences.
Why Do Our Genes Influence Political Traits?
Since the mapping of the
human genome, there is a growing body of evidence supporting the thesis
that certain inherited genetic markers are the product of human’s
evolutionary need to overcome problems related to group living (Lockyer & Hatemi, 2014; Lopez et al., 2011; Petersen, 2012, 2015).
Long before empirical approaches became available to identify and
quantify genetic influences, however, scholars approached the question
of the relationship between genes and political traits by using
deductive reasoning and evolutionary theory (Barkow et al., 1992; Cosmides & Tooby, 1997; Faulkner, Schaller, Park, & Duncan, 2004; Hammond & Axelrod, 2006; Hibbing & Smith, 2007; Kurzban & Leary, 2001; McDermott, 2004; Tooby & Cosmides, 1988, 2010; Wrangham, 1999).
They sought to explain why political attitudes and behavior appear
universally across human societies and why humans’ DNA evolved to have
such traits at all.
The empirical record shows
that all human societies revolve around communal living. Communities
provide many advantages to individuals, including greater security,
safety, support, and a division of labor between rearing children and
gathering food. Communal living, however, also creates problems that
early humans would have needed to solve, such as ensuring that everyone
has an opportunity to pass their genes on to the next generation and
that communal resources are fairly distributed. There is no one
universal solution to these problems. Humans have proven themselves to
be immensely malleable and have organized communities ranging from
strict hierarchical structures to more egalitarian pluralistic
collectives. In an evolutionary sense, the specific political solutions
were not universal: Hierarchical communities have been equally as
successful as egalitarian ones, but everyone within the tribe must
subscribe to the same solution.
This is the evolutionary
reason why people have evolved to have such strong political attitudes.
Political attitudes are different from attitudes in general. Political
attitudes “are not just how an individual feels about something but how
individuals believe others in society ought to feel and behave” (Lockyer & Hatemi, 2014,
p. 552). A person may have an attitude toward ice cream or music
without any expectation that other people ought to possess the same
attitude. The same is not generally true regarding political issues,
such as abortion, punishment for norms violators, marriage, immigration,
or how resources are allocated.
Humans have been shown to
have particularly strong political attitudes on the domains of sex and
reproduction, child-rearing, in-group/out-group relations, defense,
fairness, survival, cooperation, security, and affiliation (Apicella, Marlowe, Fowler, & Christakis, 2012; Barkow et al., 1992; Cosmides & Tooby, 1997; Fowler & Schreiber, 2008;
(p. 286)
Hatemi & McDermott, 2011a; Hibbing & Smith, 2007; Kurzban & Leary, 2001; Lockyer & Hatemi, 2014; Lopez & McDermott, 2012; Lopez et al., 2011; McDermott & Hatemi, 2013; Miller, 2011; Petersen, 2012; Petersen & Aaroe, 2012; Petersen, Sell, Tooby, & Cosmides, 2012; Smith et al., 2007; Thayer, 2000; Tooby & Cosmides, 1988, 2010; Tuschman, 2013).
People inherit the propensity to have political attitudes, but cultural
and environmental factors will shape how these political attitudes are
defined in an almost infinite number of ways. For instance,
historically, immigration has been an emotionally charged political
issue across time and between countries. The issue of “foreigners” was
as politically charged in ancient Rome as it is in modern-day Britain.
However, individuals can link immigration back to either the domain of
“out-group” relations or “fairness.” Either way, individuals are likely
to feel particularly emotional about the political issue, but they may
well fall on different sides of the debate. The context, labels, and
policies will change between countries and across time, but the most
politically charged debates will often be those that can be traced back
to one of the domains people have evolved to be the most important to
group living.
From an evolutionary
perspective, observed preferences and behaviors are not simply the
result of the current environment. Like the rest of the human anatomy,
the brain is a product of evolution. Over the course of thousands of
generations, the human brain has evolved certain universal structures
and functions to overcome recurring social dilemmas. That is, the
architecture of the human brain produces certain universal systems,
including those of cognition, emotion, preference formation, and need
for security. These evolutionary structures are different from instincts
in that they are cognitive and emotional pathways for processing
information. That is, instincts are evolved reactions to stimuli, such
as the fight-or-flight response to being surprised. Human behavioral
predispositions, on the other hand, are the product of the human brain
having evolved to process information in particular ways. For example,
humans have a strong predisposition to the in-group compared to the
out-group. However, how these groups are defined is a social contract
and is completely malleable between individuals and over time.
In this view, modern-day
political choices are more complex versions of the basic problems that
have surrounded group living since the inception of humanity. Modern
national defense and foreign policy issues and decisions to use military
force are only recent reflections of tribal decision to protect the
in-group from outsiders. The political and technological context may be
dramatically different; however, how people frame the decision and the
thought process that goes into the decision-making are the product of
thousands of years of evolution. Today’s political attitudes surrounding
child-rearing, sexual liberties, and marriage are the modern-day
equivalent of the ancestral need to ensure access to mates and have
offspring. The strong opinions on immigration reflect modern
insecurities, but those insecurities are seated in the primitive need to
balance protecting against unknown and unfamiliar others with the
potential gains from diversity (e.g., genetic diversity[strengthening
the mating pool], resource security, social norms differences, and
pathogen protections). The economic issues of today (e.g., taxes and
welfare) address the same core concerns of how to share resources in a
community.
(p. 287)
Certainly, the issues of today are more complex.
Institutions, political parties, social groups, nation states and
non-state actors, cultural and historical conflicts, elites,
communication, transportation, technology, and globalism, among many
other factors, make social interactions much more complicated than those
our ancestors confronted. Also, the relative importance and nuances of
political issues differ depending on different environmental conditions,
social forces (e.g., social customs, traditions, and material
constraints), local histories, and ecology. Nevertheless, the
fundamental psychological processes used to make political choices today
are not so different as they were in the Pleistocene. The labels and
rhetoric of issues change across cultures, time periods, and countries;
modern economic and political structures levy their own influence on the
mass public’s understanding and organization of political thoughts; and
the channels through which preferences are conveyed in large modern
societies are much less personal and direct than in the past; however,
the underlying issues that are important, including family,
reproduction, defense, and resources, remain the same. That is, when
humans have to make decisions on how to manage societal life today, they
are relying on the same cognitive and emotional mechanisms to address
the same fundamental problems that their ancestors relied on.
Genetics has brought with it a
unifying theory of human behavior—evolution. The natural, medical, and
physical sciences have a variety of microtheories, but all behavioral
studies rest on a theory of evolution. Whether or not evolutionary
theory becomes the core theory of behavior in the social sciences
remains to be seen; however, due to its capacity to lead to new
discoveries, advances, and scientific progress, it may be the vehicle by
which the social sciences catch up to the natural sciences.
Identifying Genetic Influences
A second research agenda
emerged in conjunction with the first, albeit more slowly, and it is
empirical in nature. This research approach has sought to reveal what
political traits are genetically influenced. Utilizing a classical twin
design (CTD), Lindon Eaves and Hans Eysenck (1974)
performed the first study to find that political traits are genetically
influenced. The CTD compares the co-twin correlations between
monozygotic (MZ; identical) and dizygotic (DZ; fraternal) twins. MZ
twins develop from a single fertilized egg and, thus, share their
chromosomal sequence. DZ twins grow from two separate eggs, fertilized
from two different sperm, and share approximately half their genetic
inherence, similar to non-twin siblings. This combination of twin pairs
provides researchers with the ability to control for comparable familial
influence. The underlining assumption is that, on average, MZ and DZ
co-twins grow up exposed to equivalent familial experiences, such as
having the same meals, living in the same neighborhood and at the same
time, attending the same schools, and having similar peer groups. This
allows researchers the ability to partition out factors common to the
twins (genes and familial factors) and unique to the twins (personal
experiences).1
(p. 288)
Heritability estimates focus on individual differences;
that is, they do not explain the value of a trait but, rather, the
difference of values within a population. Thus, when a heritability
estimate of 0.35 is reported for sex attitudes, for example, it is not
that genes explain 35% of sex attitudes; rather, it is that 0.35 of the
variance, or individual differences in sex attitudes within the
population, is accounted for by the aggregate of genetic influences.
That is, heritability estimates provide a population estimate of how
people differ. They are not an estimate of the percentage within any
given individual that is accounted for by genetic factors. They are not
to be interpreted to mean that for every person in the population, 0.35
of a person’s attitudes are due to genes.
Eaves and Eysenck (1974)
measured social and political attitudes on a wide range of different
issues, including the death penalty, unions, unemployment, and abortion.
Their pioneering study supported the proposition that individual
differences in political attitudes were a function of both biology and
environment. Despite their work being published in Nature, it was
more than a decade before such findings received much attention.
Indeed, the work most often credited as the foundational piece in the
genetics of politics literature was published 12 years later in Proceedings of the National Academy of Sciences. Along with his advisor, Lindon Eaves, Martin et al. (1986)
substantially extended the earlier work with much larger samples, a
wider range of political orientations, and more complex analyses that
included the role of assortative mating. This work found that
differences in political orientations were genetically influenced to a
larger extend that previously thought.
Figure 14.1 Sources of individual differences on political traits.
The first genetically informed research to make a serious impression on the social sciences, however, came much later, in 2005, when John Hibbing and John Alford (Alford, Funk, & Hibbing, 2005) republished results from Eaves and Martin’s works (Eaves et al., 1997; Eaves & Eysenck, 1974; Eaves, Eysenck, & Martin, 1989; Eaves et al., 1999; Martin et al., 1986) in the American Political Science Review. Following Alford et al. (2005), twin studies emerged as the most popular methods for exploring genetic influences on political beliefs and social behaviors. Taking up the mantle from Eaves and Martin, Hatemi, Fowler, and Dawes have perhaps led the second wave of research to rediscover the usefulness of twin studies in the research on political traits (Cesarini et al., 2008, 2009; Cranmer & Dawes, 2012; Dawes et al., 2014; Dawes & Fowler, 2009; Fazekas & Hatemi, 2016; Fowler et al., 2008; Fowler & Dawes, 2008, 2013; Fowler et al., 2011; Hatemi, 2013; Hatemi, Alford, et al., 2009; Hatemi, Eaves, et al., 2012; Hatemi & McDermott, 2014; Hatemi, McDermott, Eaves, Kendler, & Neale, 2013; Hatemi, Medland, & Eaves, 2009; Hatemi et al., 2014; Hatemi & Verhulst, 2015; Klemmensen, Hatemi, Hobolt, Petersen, et al., 2012; Klemmensen, Hatemi, Hobolt, Skytthe, et al., 2012; Littvay et al., 2011; Loewen & Dawes, 2012; Loewen et al., 2013; McDermott et al., 2013; Oskarsson et al., 2012; Smith et al., 2012; Smith & Hatemi, 2013; Verhulst, Eaves, & Hatemi, 2012; Verhulst & Hatemi, 2013; Verhulst, Hatemi, & Eaves, 2012). Many critical discoveries have come from these works, summarized in several review articles (Hatemi, Dawes, et al., 2011; Hatemi & McDermott, 2012a, 2012b, 2016; Hibbing, Smith, Peterson, & Feher, 2014) (p. 289)
First, the heritability of political traits differed greatly within and across traits (Figure 14.1
provides a summary of results from published articles). On the higher
end of the scale, roughly 50% or more of the variation in political
knowledge, social trust, ideology, participation, and interest can be
attributed to genetic influences, comparable to cognitive ability,
perceptual accuracy, and prosociality (58–70%). More than 40% of the
variation in attitudes on sex topics (e.g., gay rights), religious items
(e.g., Bible truth), economic items (e.g., welfare), defense items,
freedom and liberties, and efficacy can be attributed to genetic
influences, whereas less than 30% of the variation regarding out-group
and punishment attitudes can be attributed to genetic influences. On the
low end are ethnocentrism ( < 20%), civic duty ( < 15%), and party identification ( ~ 5%). These findings have been replicated across multiple countries and decades. For example, <20 a="" across="" and="" been="" civic="" countries="" decades.="" duty="" example="" findings="" for="" have="" href="http://sci-hub.tw/view/10.1093/oxfordhb/9780190299323.001.0001/oxfordhb-9780190299323-e-43#oxfordhb-9780190299323-e-43-bibItem-696" id="ref_oxfordhb-9780190299323-e-43-bibItem-696" identification="" multiple="" party="" replicated="" these="">Hatemi and colleagues (2014)included samples spanning from the 1970s through 2010 from Australia,
Denmark, Sweden, and the United States and found that genetic influences
on political attitudes and social, economic, and defense ideologies, in
addition to authoritarianism, were similar across time and cultures,
whereas social and environmental factors manifested themselves
differently according to local ecologies and constraints. Second,
genetic influences remained even when the most sophisticated modeling
techniques were used, such as those that included all types of
relatives; addressed the effects of assortative mating; included
specific environmental or socialization factors, or used multivariate
models, which included other psychological traits.
Assortative mating is
important to consider because both heritability and social transmission
approaches, for example, assume that mating is randomized. This is
(p. 290)
critical because studies have shown that one of the strongest correlates of long-term mates is political ideology (Alford, Hatemi, Hibbing, Martin, & Eaves, 2011; Eaves & Hatemi, 2008; Eaves, Hatemi, Heath, & Martin, 2011; Klofstad, McDermott, & Hatemi, 2012, 2013; Martin et al., 1986).
That is, even controlling for ideological convergence and social
homogamy, research has shown that people tend to select mates who share
their positions on political orientations and issues (Luo & Klohnen, 2005; Luo & Zhang, 2009).
If there is assortative mating on a trait of interest and that trait is
genetically influenced, then the inherited genetic influences will be
underestimated, while environmental and shared experiences will be
overestimated. This is true because if parents assort on the trait of
interest and the trait is genetically influenced, parents will be more
genetically similar for that trait; when they produce offspring, the
shuffling of genetic code produced by sexual reproduction will result in
less genetic variation among DZ twins (or any full sibling types) than
assumed. Thus, DZ genetic similarity will be on average greater than the
0.5 assumed in twin models, leaving MZ/DZ co-twin observed trait
differences to be explained by a smaller amount of genetic differences.
Such a circumstance would mean twin models underestimate genetic
influences. Extrapolating this to a population, the higher the
proportion of mates who share genes for a trait, the closer the DZ
correlation will become to the MZ correlation and the more the genetic
variance of this trait will be underestimated.
This means that the influence
of genetic factors in determining the political attitudes of
individuals begins even before fertilization. Hatemi and colleagues (Eaves & Hatemi, 2008; Eaves et al., 2011; Hatemi et al., 2010; Keller et al., 2009)
addressed this issue by using extended kinship models based on data
collected from twins’ parents and twins’ spouses. This technique
resulted in genetic influences accounting for even more of the variance
reported on political attitudes and ideologies. Assortative mating may
well be the single most important factor in determining an individual’s
political attitudes because it precludes and determines genetic
transmission, parent–child socialization, environment, and personal
experiences (Eaves & Hatemi, 2008).
A number of studies have also
explored the importance of specific environments. Littvay and Fazekas
included twin-specific and shared environments and also
election-specific factors, whereas Hatemi focused on critical life
events (Fazekas & Hatemi, 2016; Fazekas & Littvay, 2015; Hatemi, 2013; Littvay, 2012).
These researchers found that social factors have an important role in
shifting the relative import of both genetic and environmental
influences. When social forces (i.e., social and material constraints)
are absent, genetic influences emerge as more important in how people
differ. When personal life events or social forces are overwhelming or
constraining, genetic influences become all but absent in explaining
variation. For example, genetic influences on individual behavior will
be visible in highly disciplined religious communities. Critically
important, however, is that when the constraining crisis recedes, or
social influences allow people to differ, individuals revert back to
their initial disposition and genetic influence re-emerge as important.
One of the first substantive
challenges to the research stream on what political traits are
genetically influenced was the idea that the heritability of political
orientations
(p. 291)
might simply be an artifact or spurious, deriving from
some covariate, particularly personality. On the surface, this idea
seemed plausible; individual differences in personality are an equal
function of genes and environment. However, despite much discussion and
the widespread belief that personality is strongly related to political
traits, the empirical record shows otherwise. Most personality traits
are neither consistently nor significantly related to political values.
The major exceptions are that the personality trait of “openness to new
experiences” is positively associated with socially liberal attitudes,
whereas “conscientiousness” is positively associated with socially
conservative attitudes. Even here, the correlations between the two are
modest. Nevertheless, a series of multivariate genetic studies have
found that the overwhelming majority of genetic influences on political
traits are unique to the traits themselves and not accounted for by
personality, morality, or other psychological constructs (Dawes et al., 2014; Hatemi & Verhulst, 2015; Oskarsson et al., 2015; Smith, Alford, Hibbing, Martin, & Hatemi, 2017; Verhulst, Eaves, et al., 2012; Verhulst, Hatemi, & Martin, 2010).
There are of course many
challenges to the methods and assumptions of twin studies. We refer
readers to published critiques and exchanges for greater details
(Alford, Funk, & Hibbing, 2008a, 2008b; Charney, 2008a, 2008b, 2012; Fowler & Dawes, 2013; Hannagan & Hatemi, 2008; Hatemi et al., 2010; Hibbing et al., 2013; Littvay, 2012).
We simplify the criticisms to one dimension here. The concern largely
revolves around reductionism in some form or another. Some of the
loudest critics argue that one cannot assume general similarity in
environments, make on-average estimates, partition any trait variance
into a discrete amount of genes or environments, and so on with or
without bounds or confidence intervals. These criticisms, although they
seem reasonable, are criticisms of science in general and are specious.
In truth, no credible scientist actually challenges such concerns on a
theoretical level. However, science is naturally reductionist. Until a
better approach than empirical estimates come along, we are left with
estimates with confidence intervals and other bounds to graph or display
probabilities and possibilities.
The modern scientist knows that it is not valuable to simply ask if nature or nurture determines the form of a particular trait, or even how much
each factor contributes to a trait. Although these are often portrayed
by the methodologies available in quantitative behavioral genetics and
molecular biology, science has progressed to the point that it is known
that any dichotomy does not stand up to either empirical or conceptual
scrutiny. Quantifying the extent to which nature or nurture contributes
to a trait, however, provides a critical starting point because these
two classes of factors are both always present. They are interactive and
operate differentially during different developmental stages and life
experiences. Therefore, the methods used are simply a first step to
answer how our traits develop and how they are maintained over
the life course—that is, how it is that genetic, social, and all other
factors, including people and populations, interact to produce
behavioral traits.
Unfortunately, there is no
single method to model the complexities of life—that is, the “life”
package in R does not yet exist to perfectly model development, nor do
we have the data to do so. However, a number of extensions to classical
twin approaches
(p. 292)
have begun this process. Longitudinal twin studies have
the advantage of being able to track sources of variation across
different life stages. These studies have found that the political
attitudes of MZ and DZ co-twins were no different until they left home (Eaves et al., 1997; Hatemi, Funk, et al., 2009).
That is, in childhood, there were no genetic influences present on
political traits. Yet, when children left home, MZ co-twins continued to
show strong correlations, whereas DZ co-twins’ political attitudes
diverged. This suggested that after leaving the parental home, DZ
co-twins selected into or found themselves experiencing different
environments and social influences at greater rates than did MZ
co-twins. Thus, the home environment repressed individual differences,
and genetic influences only emerged once children left home. These
developmental studies have only begun, and there are few methodologies
and data to fully explore the developmental trajectories of genetic and
environmental influences.
How Do Genes Influence Political Traits?
Although twin studies provide
valuable insights into the latent causal genetic pathways, they cannot
identify the specific genetic and neurobiological markers and associated
systems that operate on political traits. Thus, molecular genetics has
become an emerging area of research that attempts to identify how genes
influence traits through their neurological and neurochemical pathways.
In contrast with the first empirical avenue of research that is largely
focused on revealing if and to what extent latent genetic factors
influence individual differences in political and social traits, this
avenue of research is concerned with process tracking and identifying
the specific genetic variants, how they work, and how social and
environmental factors influence and are influenced by these traits,
directly and indirectly. (For a detailed review on how genes operate
with specific pathways for political traits, see Bergner & Hatemi, 2016; Hatemi, Byrne, et al., 2012).
This avenue of inquiry has
only recently begun and thus there are only a handful of empirical
studies. To date, two approaches have been used. The first method is an
inductive empirical approach. This method involves scanning the entire
genome for a genetic marker or chromosomal region that shows a
correlation with the political trait of interest (Hirschhorn, 2009; Rietveld et al., 2013).
Like most inductive studies, this method’s strength lies in there being
no prior assumptions made on the underlying causes. That is, no prior
knowledge is assumed on the underlying biological pathways or causes of a
political trait. This avenue is in fact one of true discovery because
it identifies causal pathways that were previously unknown. Every region
of the genome is tested, and strict statistical significance thresholds
are maintained to allow for the multiple testing of the millions of
genetic markers. Being genome-wide, this method is the most empirically
rigorous of the two methods, and before any general conclusion can be
reached, independent replication is required. For example, in the first
genome-wide linkage study of political attitudes, Hatemi, Gillespie, et al. (2011) used a sample of 13,000 Australians and identified several regions containing methyl-d-aspartate, serotonin, glutamate, dopamine, olfactory, and G protein-coupled-related receptors that potentially corresponded
(p. 293)
to liberal and conservative political beliefs. However,
two follow-up studies applied genome-wide association analyses and
found no markers related to political ideologies or attitudes that
reached a significance level of 5 × 10–0.8 or better (Benjamin et al., 2012; Hatemi et al., 2014). That is, the markers were not replicated in independent samples.
The challenge of genome-wide
analyses is that they require extremely large samples in order to obtain
significant results. This is because the influence of any single
genetic marker on a complex social or political trait, such as vote
choice or ideology, is infinitesimal. Indeed, Rietveld et al. (2013)
showed that for complex traits such as educational attainment, samples
into the hundreds of thousands are required to identify individual
genetic markers that are statistically significant. As such, although
genome-wide analysis is a highly promising avenue of research, there is
is currently an insufficient amount of data to accurately perform this
analysis. Thus, technological and methodological advances and additional
data in the future are bound to turn this avenue of research into one
of the most important in the field.
A second gene mapping
technique relies on a priori knowledge to identify likely candidate
genes—that is, those genes that can be expected to be associated with
specific political and social traits based on prior associations with
similar or root traits. The “candidate marker approach” has proven to be
more popular than genome-wide studies. This should not be surprising
because it reduces costs, saves time, and exploits the greater
availability of data. Fowler and Dawes (2008)
relied on previous evidence that identified two genetic polymorphisms,
monoamine oxidase A (MAOA) and 5HTT (serotonin), as being associated
with prosocial and antisocial behavior. MAOA, for example, correlated
with prosocial behavior, and a lack of MAOA appeared to be significant
in antisocial behavior. Treating voter turnout as a prosocial behavior,
they found that the “high” allele of MAOA and the “long” allele of 5HTT
were related to higher voter turnout. However, these markers were only
significant in specific socializing conditions, such as religious
service attendance. That is, genetic markers previously associated with
sociability required some additional social priming in order to operate
on voter turnout. Several follow-up articles presented some evidence of
replication for serotonin having a role in participation (Dawes & Fowler, 2009; Deppe, Stoltenberg, Smith, & Hibbing, 2013; Fowler & Dawes, 2008, 2013).
Although there have been some
hints at novel pathways using candidate gene studies, most of these
studies account for very small variance and few have stood up to
replication (Duncan & Keller, 2011).
At this early stage, no single genetic variant has been found to
account for a significant portion of a discrete political trait with a
strong degree of certainty. The lack of clear associations is partly a
reflection of the fact that this area of genetic research remains in its
infancy.
These findings, however, are
only a first step. Research has begun to link differences in the gene
sequence, genetic expression, hormones, and neurological function, which
will lead to a better understanding of the full developmental and
biobehavioral pathways to political preferences and actions.
Endocrinological, psychopharmacological, and neurochemical studies have
observed the role of hormones in regulating the release and modulation
of peptides (e.g., vasopressin, serotonin, dopamine, and oxytocin),
which in turn
(p. 294)
alter neurological function identified through
event-related brain potentials. Functional magnetic resonance imagining
and lesion studies show that activity in the brain is triggered by
subcellular activity and specific genetic mechanisms under specific
environmental conditions. This research has been conducted on a wide
range of traits correlated with political traits, including emotion
regulation and recognition (Canli & Lesch, 2007; Uzefovsky, Shalev, Israel, Knafo, & Ebstein, 2012), social affiliation (Walum et al., 2012), cognition (Meyer-Lindenberg et al., 2006), anxiety (Myers et al., 2014), moral judgments (Koenigs et al., 2007), power seeking (Madsen, 1985, 1987), stereotyping of out-groups and in-groups (Hart et al., 2000), trust (Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, 2005), self-awareness (Gusnard, 2005), empathy (Carr, Iacoboni, Dubeau, Mazziotta, & Lenzi, 2003), and decision-making (Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003).
What Have We Learned?
Perhaps the greatest
challenges to the integration of genetics into the social sciences are
dispelling myths and the need to find a common language. The vast
majority of degree programs in the social sciences do not require basic
science courses; thus, social and natural scientists start with large
knowledge gaps. Most PhD graduates in social sciences have only a basic
knowledge of how genes operate and, as such, many interpret genetic
research as arguing that human social behavior is purely deterministic,
which is absurd. On the other hand, those entirely focused on genetics
or biology often miss the subtleties of the social constructs they seek
to understand. For instance, neuroimaging, physiological, and other
biological studies have often mistakenly treated party identification,
ideology, attitudes, and vote choice as synonyms when, in fact, they are
well known to be quite different constructs. Party identification, for
example, is more often about identity than values, whereas discrete
attitudes tend to reflect beliefs. So, for the social scientist
unfamiliar with genetics, what are the takeaways? What have we learned
in the past 40 years at the intersection of genetics and politics?
First, political traits are
not simply the result of modern institutions, governments, economies, or
circumstance. Rather, humans have evolved to be political in order to
address the needs, constraints, and benefits that can be derived from
communal living. Politics is a fundamental part of what it is to be
human—not in a colloquial sense but in an evolutionary sense.
Second, traits result from
the interaction of both genetic and environmental influences through
developmental pathways. Genetic factors only operate in conjunction with
environmental ones. There is no “gene for” any complex trait. That is,
there is no “political” gene. No complex social trait will ever be only
nature or nurture. Genetic influences on the predilection to exhibit any
political or social trait will be indirect and result from the
aggregate effects of the interaction of thousands of genes interacting
with immeasurable immediate and long-term environmental conditions and
experiences that change during the life course.
(p. 295)
Third, genetics is as much about human universals as it
is about individual differences. Understanding the cognitive and
emotional mechanisms will allow researchers to map the basic
architecture of how humans operate, such as universal brain function,
the role of specific hormones, and the downstream cognitive and
emotional processes humans use to make decisions and form preferences.
At the same time, although all healthy people use the same neural
architecture, people differ, even if so slightly in their genetic
makeup, resulting in differences within those cognitive, emotive, and
perceptual pathways, which will result in trait variation. People are
different. We are different at the genetic level, and we are different
socially. Environmental forces will not have the same impact on all
individuals. Genetic differences will not have the same result across
environments. For example, only some soldiers from the same unit who
fight in the same battle develop post-traumatic stress disorder; others
repress the experience, and still others write books and easily recount
the battle’s details. Even when raised in the exact same environments,
there will be differences in individual responses to stimuli. The
overwhelming majority of research on political traits treats people as
the same (rational actors) and views socializing agents and experiences
as the source of individual differences. This simply cannot continue
under the weight of the evidence to the contrary. The hundreds of
millions of combinations of polymorphic markers that differ between
people in conjunction with social experiences lead to differences in
perception, cognition, emotion, reasoning, preferences, and eventual
behavior. In short, the one tenet that buttresses many of our core
theories in social science, the blank slate, is no longer valid.
Fourth, although there are
major individual differences within societies, there are scant
differences across them. People are different. Peoples are the same.
Less than 1% of DNA differs between individual humans (Redon et al., 2006), and more than 85% of those differences exist within populations (Jorde & Wooding, 2004).
In other words, there are greater genetic differences door-to-door than
continent-to-continent. Molecular genetic research affirms that we are
mostly the same across ethnicities. This simple fact has arguably led to
increases in tolerance and changes in policy. We can think of no better
or more politically salient example than the issue of equal rights for
lesbian, gay, bisexual, and transgender (LGBT) individuals in Western
democracies. Since Hamer, Hu, Magnuson, Hu, and Pattatucci (1993)
implicated genetic loci for homosexuality, the narrative changed from
one of deviant behavior and choice to one of inherent disposition
(Hatemi & McDermott, 2011c, 2012a).
Subsequently, elite views, including state and US Supreme Court
decisions, changed, which filtered down to the public’s change in its
view of sexual preference. Today, the trend is acceptance and increasing
tolerance.
Fifth, identifying genetic
systems and their related perceptual, physical, cognitive, and emotional
processes has led to novel hypotheses regarding the nature and
manifestation of political preferences and behavior. For example, we
have learned that smell and pathogen avoidance, and their genetic
mechanisms, have a significant role in determining why individuals
differ in their attitudes about immigration (Hatemi et al., 2013; McDermott, Tingley, & Hatemi, 2014; Navarrete & Fessler, 2006). Such findings could not have been obtained using social- or environmental-only approaches to behavior.
(p. 296)
Finally, in a world of both genes and environments,
environments become more—not less—important. Genetic theory provides a
vast new avenue to explore environmental influences. It is not that
people are born liberal or conservative or any other political
orientation. Nor are they simply socialized as so. Genes rarely, if
ever, have a direct role in any complex trait. Rather, individuals with
certain genetic profiles will be more or less likely to find themselves
engaging in or seeking certain experiences and then perceiving and
reacting to experiences in a certain way. The sum of those interactions
alongside all the differences in constraints, opportunities, and people
with whom they engage, among others, and simple happenstance will result
in certain probabilities of, for example, joining the military, turning
out to vote, being influenced by a candidate’s speech, or even running
for office (Fazekas & Hatemi, 2016).
In this sense, sociology is even more important in explaining why
people differ. Understanding what a gene is, which genes are activated
or repressed, and how they inform behavior requires a greater, not
lesser, understanding of the social environment. By including genetics,
the environment represents much more than the observed stimuli humans
experience. Rather, the concept of the “environment” expands beyond
social experiences but includes the in utero environment, internal
cellular environments, and all the experiences that occur across the
lifespan. That is, environment includes everything inside and outside
the body both before and after an individual is born. It also includes
the environments of our ancestors. Differences in the individual’s
circumstances, such as war or having a child, can have effects on both
the internal mechanisms and the person’s overall behavior, but these
experiences can trigger entirely different genetic mechanisms that may
or may not work together or in opposition to one another. Environments
trigger and restrict gene expression, and without these triggers,
genetic influences are not realized. In short, without biology, we
cannot fully explain why people differ under the same environmental
conditions, and without the environment, we cannot explain why people
with the same DNA differ in behavior outcomes. As such, we have much
more to do to reinvest in understanding socialization, and life
experiences, than ever before.
Conclusion
Now, perhaps more than at any point in the past 200 years, the social sciences are back in a time of discovery. King’s (2011)
observation of how we viewed our future 60 years ago does not have to
be how we view our future today. We rarely have had the means of true
discovery in the social sciences. Rather, social scientists have most
often sought to empirically validate something already observed.
Genetics identifies the unobserved. It provides a starting point to
identify novel developmental pathways to preferences and behaviors.
Understanding differences in the genome can help identify why people who
experience the same social environment physically perceive it
differently and react to it differently. It allows for the exploration
of individual trajectories. This line of research
(p. 297)
has only begun, and we are excited to see genetic and
other biological methods and approaches further incorporated into the
study of political traits.
Notes:
(1.)
In addition to the classical twin design, a number of studies have
reported similar finding using twins reared apart and adoption studies (Abrahamson, Baker, & Caspi, 2002; Bouchard & Loehlin, 2001; Bouchard, Lykken, McGue, Segal, & Tellegen, 1990; Bouchard & McGue, 2003; Oskarsson et al., 2015; Tesser, 1993).
Posts
