This volume examines cutting edge research in the study of biology and politics. Following an introduction from the edit
169 86 1MB
English Pages 264 Year 2011
BIOLOGY AND POLITICS: THE CUTTING EDGE
RESEARCH IN BIOPOLITICS Series Editors: Steven A. Peterson and Albert Somit Recent Volumes: Volume Volume Volume Volume Volume Volume Volume
1: Sexual Politics and Political Feminism, 1991 2: Biopolitics in the Mainstream, 1994 3: Human Nature and Politics, 1995 4: Research in Biopolitics, 1996 5: Recent Explorations in Biology and Politics, 1997 6: Sociobiology and Politics, 1998 7: Ethnic Conflicts Explained by Ethnic Nepotism, 1999 Volume 8: Evolutionary Approaches in the Behavioral Sciences: Toward a Better Understanding of Human Nature, 2001
RESEARCH IN BIOPOLITICS VOLUME 9
BIOLOGY AND POLITICS: THE CUTTING EDGE EDITED BY
STEVEN A. PETERSON Pennsylvania State University, Harrisburg, USA
ALBERT SOMIT Southern Illinois University Carbondale, USA
United Kingdom – North America – Japan India – Malaysia – China
Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2011 Copyright r 2011 Emerald Group Publishing Limited Reprints and permission service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. No responsibility is accepted for the accuracy of information contained in the text, illustrations or advertisements. The opinions expressed in these chapters are not necessarily those of the Editor or the publisher. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-85724-579-3 ISSN: 2042-9940 (Series)
Emerald Group Publishing Limited, Howard House, Environmental Management System has been certified by ISOQAR to ISO 14001:2004 standards Awarded in recognition of Emerald’s production department’s adherence to quality systems and processes when preparing scholarly journals for print
CONTENTS LIST OF CONTRIBUTORS
vii
PART I: INTRODUCTION INTRODUCTION Albert Somit and Steven A. Peterson
3
PART II: GENES, EVOLUTION, AND POLITICS THE STATES’ RESPONSE TO PARENTAL DIVESTMENT: COULD SAFE HAVEN LEGISLATION LEAD TO MORE CHILD ABANDONMENT? Laurette T. Liesen
11
A MODEL OF ACTIONS AND NORMS: AN INTEGRATED EVOLUTIONARY PERSPECTIVE ON NORMATIVE ETHICS AND HUMAN BEHAVIOR Birgitta S. Tullberg and Jan Tullberg
29
THE BIOPOLITICS OF PRIMATES Johan M. G. van der Dennen
53
MEASURING SOCIAL AND POLITICAL PHENOTYPES Levente Littvay
97
POLITICAL SCIENCE AND BEHAVIOR GENETICS: RETHINKING FOUNDATIONAL ASSUMPTIONS Evan Charney v
115
vi
CONTENTS
FROM GENES TO POLITICS: BRIDGING THE GAP BETWEEN BIOLOGICAL AND SOCIAL EXPLANATIONS OF POLITICAL BEHAVIOR VIA TWIN STUDIES Rebecca J. Hannagan
139
GENES, TWIN STUDIES, AND ANTISOCIAL BEHAVIOR Danielle Boisvert and Jamie Vaske
159
PART III: THE BRAIN AND POLITICAL BEHAVIOR NEUROLOGICAL IMAGING AND THE EVALUATION OF COMPETING THEORIES Dustin Tingley
187
BRAIN SCIENCES AND POLITICS: SOME LINKAGES Robert H. Blank
205
BRAIN IMAGING AND POLITICAL BEHAVIOR: A SURVEY John M. Friend and Bradley A. Thayer
231
LIST OF CONTRIBUTORS Robert H. Blank
University of Canterbury, Christchurch, New Zealand
Danielle Boisvert
Pennsylvania State University, Middletown, PA, USA
Evan Charney
Duke University, Durham, NC, USA
John M. Friend
University of Hawaii at Manoa, Honolulu, HI, USA
Rebecca J. Hannagan
Northern Illinois University, DeKalb, IL, USA
Laurette T. Liesen
Lewis University, Romeoville, IL, USA
Levente Littvay
Central European University, Budapest, Hungary; Visiting Scholar Washington State University, Pullman, WA, USA
Steven A. Peterson
Pennsylvania State University, Middletown, PA, USA
Albert Somit
Southern Illinois University, San Diego, CA, USA
Bradley A. Thayer
Baylor University, Waco, TX, USA
Dustin Tingley
Harvard University, Cambridge, MA, USA
Birgitta S. Tullberg
Stockholm University, Stockholm, Sweden
Jan Tullberg
Stockholm University, Stockholm, Sweden
Johan M. G. van der Dennen
Rijksuniversiteit Groningen, Groningen, Netherlands
Jamie Vaske
Western Carolina University, Cullowhee, NC, USA vii
PART I INTRODUCTION
INTRODUCTION Albert Somit and Steven A. Peterson Biopolitics is not altogether a felicitous term used to describe the approach of those political scientists who use biological concepts, with neo-Darwinian evolutionary theory at the center, and biological research techniques to study, explain, predict, and sometimes even to prescribe political phenomena. Allusions to biological influences on human politics are as old as the Greek philosophers (for more detail, see Peterson, 1976). Plato’s metaphor of bronze, silver, and gold, developed in The Republic, is an early analogue to later suggestive work on the genetic bases of human behavior. Here, Plato argued that certain people were born with the capacity to rule; most, however, were born with the more limited capacity to be ‘‘producers.’’ Aristotle referred to inherent qualities of humans as shaping their behavior. In his Politics, he noted ‘‘y that some should rule, and others be ruled is a thing not only necessary but expedient; from the hour of their birth, some are marked out for subjection, others for rule.’’ In the 1800s and early 1900s, these arguments re-emerged as racial analysis. Sir Francis Galton, cousin of Charles Darwin, attempted empirical analysis of racial differences (Galton, 1892). One approach was to determine the number of famous people or great men within a variety of different cultures; the greater the proportion of great men in a population, he contended, the more superior that culture. He concluded that the English ‘‘race’’ was among the most superior of his time. Regrettably, he concluded, the ancient Athenians were as superior over the contemporary English as the English were over Africans. Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 3–8 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009002
3
4
ALBERT SOMIT AND STEVEN A. PETERSON
A number of scholars, such as Charles Louis Secondat and Baron de Montesquieu, posited a linkage between climate and group differences. He noted for instance, the following difference between hotter and cooler climes: ‘‘The people of hot countries are [timid] as older men are: those of cooler countries are more courageous as are the young people’’ (Montesquieu, 1955, p. 189 [author’s translation]). Other scholars who advanced a similar thesis included Raymond Gettell (1933) and Heinrich von Treitschke (1916). Historically, other thinkers used an organismic metaphor to understand the political realm. John of Salisbury, in Policraticus (1159), argued that society was like a biological organism. The commonwealth was the body, with the king as its head, the church as its soul, and all other members of the body politics performing lesser functions. Thomas Hobbes, of course, put an organismic metaphor front and center in his Leviathan. He referred to the state as an artificial animal. He likened sovereignty to the soul, magistrates to joints, reward and punishment to nerves, wealth and riches of members of the state to strength, and so on. Herbert Spencer later on used a clearly organismic analogy, as did two early key figures in the development of American political science – A. Lawrence Lowell and Woodrow Wilson. In his Constitutional Government, for instance, Wilson noted that ‘‘y government is not a machine but a living thing. It falls, not under the theory of the universe, but under the theory of organic life.’’ Significantly, he goes on to state that ‘‘It is accountable to Darwin, not to Newton’’ (Wilson, 1908, p. 56). Finally, some analysts determined that there were public policy implications from racial or evolutionary analyses. John W. Burgess, another key figure in the evolution of American political science, argued that the Teutonic people were uniquely qualified to produce political societies of eminence. One deduction from this for Burgess was that the Teutonic societies would have to carry their genius to other parts of the world where people were incapable of creating advanced civilization (Burgess, 1890). Thus, racial analysis led to the argument for colonial expansionism. Madison Grant perhaps took this perspective to its starkest conclusion. In the introduction to the fourth edition of his work, The Passing of the Great Race, Grant argued for ‘‘y the decision of Congress of the United States to adopt discriminatory and restrictive measures against the immigration of undesirable races and peoples’’ (Grant, 1923, p. xxviii). In short, his sense that there was differentiation by race in the quality of people led him to argue for official discrimination against the ‘‘lower’’ sort. The contemporary interest in biopolitics can be traced most directly to the 1960s. Several different events/publications during this decade serve as the
Introduction
5
baseline for subsequent developments. In 1964, Lynton Caldwell published a piece in the Yale Review on environmental policy (Caldwell, 1964). In 1963, James C. Davies published his book, Human Nature in Politics, a work that suggested biological components to human political behavior. Finally, in 1967, at a meeting of the Southern Political Science Association in New Orleans, Albert Somit (1968) and Robert Pranger (1967) presented papers on biological bases of political behavior. Which of these starting points one might embrace is really beside the point. It is clear that the 1960s marked the beginning of an interest in the linkage between biology and politics. At the heart of biopolitics is evolutionary theory. This is the intellectual core. Scholars in this field begin with the assumption that human behavior, including human political behavior, is a product of the evolutionary process. Thus, to understand key elements of human politics, we need to understand evolution. Currently, the reigning paradigm is ‘‘neo-Darwinian theory.’’ This combines Darwin’s insights on natural selection and the body of knowledge from genetics. The two work together to provide a powerful paradigm for explaining evolutionary change. The basic components of Darwin’s theory are deceivingly simple: 1. More offspring are born than can survive. This creates a struggle for existence, given the limitations on population that any environment can support. 2. There is variation among organisms. Some of these variations facilitate increased odds of survival and subsequent reproductive success; other variations may be more likely to lead to death and reduce the chances for reproductive success. 3. The variations that lead to increased chances of survival and reproductive success are transmitted from parent to offspring. The process by which some individuals survive and reproduce and pass on their characteristics to the next generation is natural selection. Some animals are ‘‘selected’’ for reproductive success, whereas others are not. Those characteristics leading to reproductive success are passed down through generations. Darwin himself could not account for the process by which variations that were selected were passed from one generation to the next. Genetics provides the answer to that. Genes are the messengers between generations. An organism’s structure and functioning is the result of the genetic structure in interaction with the larger environment. Natural selection selects for
6
ALBERT SOMIT AND STEVEN A. PETERSON
certain genes, as these genetic ensembles increase the odds of later reproductive success of the organism. More recently, with the publication of Edward O. Wilson’s Sociobiology (1975), neo-Darwinian theory has been applied to social behavior. Recall that an individual should behave in such a way as to increase the number of his or her genes in the next generation. There are two avenues toward this: first, by passing along one’s genes directly, usually referred to as ‘‘individual reproductive success’’ and, second, one can behave in such a manner as to increase the reproductive success of one’s relatives, with whom one shares genes (Barash, 1982). The combination of individual reproductive success and that of relatives is termed ‘‘inclusive fitness,’’ a key concept in sociobiology. In any species, accordingly, the individual should behave in such a way as to maximize the number of his/her genes appearing in the next generation. This includes social behavior, and human political behavior is one species of social behavior. Thus, applying Darwinian theory through the specific theoretical perspective of sociobiology, many in biopolitics have explored the impact of inclusive fitness on human political behavior (e.g., see Somit & Peterson, 1997). One of the central issues that distinguishes those in biopolitics from mainstream political science is the emphasis on the need to understand the evolutionary origins of the behavioral predispositions which Homo sapiens share with all the other social primates and which significantly affect our political life. Profoundly influenced by ethology and contemporary evolutionary theory, those working in this area insist that we should give proper weight to the role played by nature as well as by nurture in shaping our social and political behavior. (1) That insistence runs directly counter, of course, to the long-accepted disciplinary paradigm, which holds that human political behavior is learned and that possible genetically transmitted proclivities can and should be ignored. (2) Closely related to this thrust, as we shall see, is another though much smaller group, which seeks to derive ethical and moral guidelines for political decisions from evolutionary theory, in effect, to rest a political philosophy on a scientific basis. This volume is subtitled ‘‘The Cutting Edge’’ to reflect coverage of some of the latest developments in the study of biology and politics. One of the two most obvious examples are sections on genes and politics, specifically, the use of twin studies data to assess genetic contributions to politically relevant behavior. The other constitutes the recent efforts to use knowledge and instrumentation from the brain sciences to the study of politics.
Introduction
7
In Part II of the book, we begin with three general essays on evolution and politics, to illustrate the state of the art therein today. Laurette Liesen uses evolutionary theory as the basis for exploring the adequacy of abandoned baby legislation. Her use of evolution adds an interesting and provocative twist to the policy debate. Birgitta Tullberg and Jan Tullberg use an evolutionary approach to address issues of normative ethics in human behavior. Finally, Johan van der Dennen does a great service by cataloging many articles on primate politics. Given the close relationship of humans with their primate relatives, this detailed essay provides the building blocks for a comparative analysis of human and nonhuman primate politics. The reader, then, can examine how his or her understanding of human politics is related to the political behavior of our primate cousins. Twin studies as a research methodology has recently emerged within the discipline of political science. A block of four chapters elaborates. Levente Littvay begins with measurement issues associated with twin studies research. For instance, he discusses in some detail reliability and validity. Evan Charney provides a critical perspective on behavior genetics as used in political science, as well as raising more general issues of twin studies. But what of the substance? Rebecca Hannagan provides a survey of political science literature using the twin studies approach and adds her perspective to the enterprise. Finally, Danielle Boisvert and Jamie Vaske have authored a work using twin studies to explain the genetic bases of antisocial behavior. In addition, they use data that identifies genetic locations of such behavior. This chapter illustrates ‘‘how to do it’’ for the interested reader. Given that antisocial behavior is a major focus of the political and policy process (consider the importance of law enforcement in the political system), the subject is of substantive interest as well. Part III explores the relevance of the brain science for our understanding of politics. Robert Blank capably summarizes the relevance of knowledge of brain function for the study of politics. John Friend and Bradley Thayer look more closely at methods – a survey of work on brain imaging and politics. There is a slowly increasing body of research that adopts brain imaging technology such as fMRI – functional magnetic resonance imaging) to study political phenomena. Dustin Tingley also describes the use of neurological imaging technology with a number of illustrative examples. The editors hope that this volume serves to introduce interested readers to cutting-edge issues in biopolitics and might induce some to consider adopting theory and method from this volume into their own research work. To the extent that this volume contributes to the advancement of research in biology and politics, it will have made a unique contribution.
8
ALBERT SOMIT AND STEVEN A. PETERSON
REFERENCES Barash, D. P. (1982). Sociobiology and behavior (2nd ed.). New York: Elsevier. Burgess, J. W. (1890). Political science and comparative constitutional law. Boston: Ginn. Caldwell, L. K. (1964). Biopolitics: Science, ethics, and public policy. The Yale Review, 54, 1–16. Davies, J. C. (1963). Human nature in politics. New York: Wiley. Galton, F. (1892). Hereditary genius (2nd ed.). London: MacMillan. Gettell, R. (1933). Political science. Boston: Ginn. Grant, M. (1923). The passing of the great race. New York: Scribner’s. Montesquieu, Baron de. (1955). De l’Esprit des Loix, Societe des Belles Lettres, Paris. Peterson, S. A. (1976). Biopolitics: Lessons from history. Journal of the History of the Behavioral Sciences, 12, 354–366. Pranger, R. (1967). Ethology and politics. Presented at Southern Political Science Association meeting, New Orleans. Somit, A. (1968). Toward a more biologically oriented political science. Midwest Journal of Political Science, 12, 550–567. Somit, A., & Peterson, S. A. (1997). Darwinism, dominance, and democracy. Westport, CTL: Praeger. Treitschke, H. von (1916). Politics (English translation). London: Constable. Wilson, E. O. (1975). Sociobiology. Cambridge: Harvard University Press. Wilson, W. (1908). Constitutional government in the United States. New York: Columbia University Press.
PART II GENES, EVOLUTION, AND POLITICS
THE STATES’ RESPONSE TO PARENTAL DIVESTMENT: COULD SAFE HAVEN LEGISLATION LEAD TO MORE CHILD ABANDONMENT ? Laurette T. Liesen During the autumn of 2008, dozens of parents showed up at hospitals in Nebraska to abandon their children. One father showed up at the Creighton University Medical Center emergency room with his nine children, ages 1–17 years, to abandon them so that they would be safe. Overwhelmed with caring for them alone after his wife died shortly after delivering their ninth child and having to quit his job, he claimed that he no longer could take care of his children (Ross, 2008). Nebraska was one of the last states in the United States to pass a safe haven law to prevent the abandonment and death of infants. Its law took effect in July 2008, but it did not specify the age of a child a parent could abandon without prosecution. Instead, the law simply stated that children could be abandoned at hospitals by parents who wanted to abandon them. Because the law was vague in regard to the age limit, parents from around Nebraska and other states began abandoning their preadolescent and adolescent children. In fact, one mother drove from Georgia to abandon her child because of his violent and uncontrollable behavior (Eckholm, 2008a). By the time the Nebraska legislature met in November 2008, 35 children, mostly adolescents, were abandoned in Nebraska. Of the 35 that were left by their parents, 27 of them received Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 11–27 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009003
11
12
LAURETTE T. LIESEN
mental-health services, 28 came from single-parent homes, and 22 had a parent/guardian in jail. In terms of race, 20 were white, 9 were AfricanAmerican, and 1 was Native American (Slevin, 2008). Because of the consequences of its initially vague safe haven law, Nebraska’s legislature held a special session in November 2008 to change the age limit to 30 days old for those parents wishing to abandon their infants without prosecution (Eckholm, 2008b). Establishing laws that allow parents to abandon their babies without prosecution and with anonymity have not only swept the United States, but other countries as well. Both Italy and Japan have set up modern mechanisms for parents to anonymously abandon their infants. In Kumamoto, Japan Jikei Hospital set up a drop off for infants called the Stork’s Cradle. Controversy arose when a parent dropped his 3-year-old toddler in the hospital’s hatch that leads to an incubator (Greimel, 2007). After finding 30 abandoned babies over a two-year period, the Casilino Polyclinic in Rome set up a small structure that has a heated cradle along with a respirator and other life saving instruments. Other modern ‘‘foundling wheels’’ also have been set up in Germany, Switzerland, the Czech Republic, and other European countries (Povoledo, 2007). In the United States over the past 10 years, all 50 states have passed their own versions of safe haven laws, enabling parents to anonymously abandon their infants at various ages at hospitals or police/fire stations without penalty. These laws attempt to provide a means for desperate mothers to abandon their infants safely. While most states passed these law based on little research (Baran, 2003), much of the current research on these laws examines the legal, psychological, and social aspects of this policy area. None so far incorporate evolutionary perspectives in their discussions of safe haven laws that would provide an explanation into the motivations that lead mothers to abandon their children. Throughout human history, under certain conditions, and as a reproductive strategy, parents will divest from their children through abandonment or infanticide if they feel they cannot raise them, and if it is in their long-term reproductive interests to do so. At this point, safe haven laws focus on saving infants without seriously addressing the environmental variables and individual circumstances that lead parents to decide to abandon their children, and an evolutionary perspective can fill that gap. Based on historical and cross-cultural analyses and insights from evolutionary theory, this chapter argues that safe haven laws may increase child abandonment and encourage ambivalent parents to terminate their parental investment in infants and even older children by providing legally and socially acceptable mechanisms to do so.
The States’ Response to Parental Divestment
13
OVERVIEW OF STATES’ SAFE HAVEN LAWS Safe haven laws arose as a compassionate response to the perceived increase in the number of mothers who killed their infants or abandoned them in unsafe places, such a dumpsters, toilets, outdoors, etc. (Appell, 2002b; Sanger, 2006). The policy problem of infant abandonment arrived on the local policy agenda in Mobile, Alabama in 1997 and early 1998. During that time, 20 infants were reported abandoned. In one case, a mother and grandmother drowned an hour-old infant in a toilet, and each received a 25-year prison sentence (Sanger, 2006). In response to this case, the program called ‘‘A Secret Safe Place for Newborns’’ was established. Prosecutors promised anonymity and immunity if the infant was relinquished unharmed. In 1999 Texas also experienced a surge in abandoned babies – 13 were abandoned in a 10-month period, 3 of whom died. Texas’ Baby Moses Law was the nation’s first safe haven law passed in 1999. Within two years, dozens of states passed safe haven laws with little debate, analysis, or opposition (Baran, 2003; Sanger, 2006). In order to reduce the occurrences of neonaticide and infanticide in which infants were left to die, all 50 states in the United States have passed safe haven laws. In 34 states, safe haven laws expressly attempt to protect the anonymity and freedom of the discarding parent, and 18 states allow someone other than the parent to relinquish an infant (Guttmacher Institute, 2010). All 50 states plus Washington, DC have designated the places or personnel authorized to accept an abandoned infant. Thirty-two states have authorized emergency services personnel to accept infants at fire or police stations, and 47 states allow hospital or health-clinic employees to accept abandoned infants. Finally, 12 states allow infants to be relinquished at adoption agencies or a specific facility designated by that state, such as a pregnancy crisis center. Twenty-five states do mandate a protocol in which the person relinquishing the infant and the infant itself wear identification bracelets to facilitate future reunification, and provide medical information about the infant. These states also investigate whether or not the child relinquished is also registered as a missing child (Guttmacher Institute, 2010). Since the safe haven laws have decriminalized the abandonment of children in specific places and redefines it as relinquishment, it enables the states to expedite the termination of parental rights and the adoption process for the abandoned child (Appell, 2002b). In terms of age limits at which an infant can be relinquished legally, the states do vary. Initially, many states only allowed infants to be abandoned up to three days old, and now only 14 states do today. Most states have
14
LAURETTE T. LIESEN
expanded the window of opportunity for abandonment in response to the continuation of illegal abandonments. States’ limits range from 5 days (New York) to 14 days (Iowa) to 30 days (New Jersey) to 90 days (New Mexico) and up to one year (North Dakota) (Guttmacher Institute, 2010). After Nebraska failed to define a limit in its first safe haven law in 2008, parents from all over the country abandoned children of various ages there, forcing the state to establish an age limit of up to 30 days. As a case study, this chapter looks specifically at Illinois’ experience with its safe haven law, and examines how it has expanded over the past nine years, giving parents additional time and places to abandon their infants legally. Within Illinois, the problem of abandoned babies landed on the policy agenda in the late 1990s. Between 1997 and 1999, 32–34 infants were abandoned each of these years in Illinois. By 2000 the number of abandoned children under the age of one increased to 38 (Safe Place for Newborns: About the Program: Facts and Stats, 2001). Like other states’ policy experiences, there was no opposition to the proposed safe haven bill. The Illinois Senate and House passed the bill unanimously, and it was signed into law in August 2001. Initially, the Illinois’ Abandoned Newborn Infant Protection Act provided immunity for mothers who abandon their infants at a hospital or fire station unharmed within 72 hours of birth. The law also stated that these infants must be turned over to the Illinois Department of Children and Family Services (DCFS) in order to start adoption procedures. At the same time, the law states that DCFS must also make an effort to locate the abandoned infant’s father. Finally, the law required that the state conduct a media campaign so that people know that these safe havens are available. Even though the information campaign is the program’s only expenditure, the fact that the implementation of the program had no costs was very appealing to legislators (Lawmakers Vote to Let Mothers Leave Newborns in Safe Places, 2001). Despite passing this law in 2001, infants in Illinois were still being abandoned illegally. In response to this persistent problem, the state continued to expand the time period and places available to parents for legal abandonment. In 2004, Illinois added police stations as a legal place to abandon infants unharmed. In 2006, Illinois extended the relinquishment period to seven days and mandated that safe haven information be taught in schools’ health-education classes. By 2007 Illinois also mandated that all Illinois safe haven sites display a sign identifying them as places where infants can be relinquished (Save Abondoned Babies Foundation, 2009). The latest changes in Illinois’ safe haven law include consolidating
The States’ Response to Parental Divestment
15
information for abandoning parents into a three-page brochure with the option of sending the state medical information (Save Abondoned Babies Foundation, 2010), and extending the period in which a parent can relinquish an infant to 30 days (Fifty-three Safe Babies– So Far, 2009). Since 2001 Illinois has managed to collect 61 legally abandoned infants, but have also found 58 illegally abandoned infants, of whom 28 had died (Save Abondoned Babies Foundation, 2010; Marrazzo, 2009). Of the 119 documented infant abandonments, 51 percent were legally abandoned. While the state governments have embraced safe haven laws as the solution for child abandonment, there are critics who question whether these laws will work or are the best solution to this problem. In a report issued by the Evan B. Donaldson Adoption Institute entitled ‘‘Unintended Consequences: ‘Safe Haven Laws’ Are Causing Problems, Not Solving Them,’’ it argues that there are several problems with these laws (Baran, 2003). First, it maintains that safe haven laws encourage women to conceal their pregnancies, leading them to give birth without medical care and perhaps in an unsafe place. In some states, individuals other than the mother have the opportunity to abandon infants without the mother’s consent. At the same time, safe haven laws have set up an outside option that operates separately from existing child-welfare programs that already deal with abused and neglected children and adoption policies. Finally, this organization maintains that the state is promoting abandonment as an easier option than the adoption process, thus sending the message that deserting infants is socially acceptable. While there seemed to be a sudden surge in child abandonment and infanticide, this is truly not the case. Both the historical evidence and evolutionary analyses show that some parents, under certain conditions, will choose to divest from their children when it is in their reproductive advantage to do so. In the past, nations that provided a socially expedient and acceptable means to do this definitely experienced an increase of infant and child abandonment.
EUROPEAN EXPERIENCES WITH CHILD ABANDONMENT Child abandonment has occurred throughout human history and across cultures. In terms of Western history, child abandonment occurred throughout Europe from the Hellenistic period until contemporary times
16
LAURETTE T. LIESEN
(Boswell, 1988). According to Fuchs (1987), during the nineteenth century, there were two models used in Europe to deal with unwanted children: one used by Britain and the other by France. In Britain unwed mothers could get financial support from the fathers of their children, and many would send these children to the countryside to be raised by rural women who were paid to care for these unwanted, and usually illegitimate, children. In these cases, the parents were financially responsible for the care of their children. The foundling homes that were established in Britain were selective as to the infants they accepted. They did not have a wheel or tour that accepted all infants, and this reduced the number of infants abandoned as well as the mortality rate compared to other parts of Europe (Fildes, 1988). In France, the fathers of illegitimate children were not held responsible and the unwed mothers usually had no resources to care for their children. In response to this problem, France and Russia established foundling home systems in order to remove unwanted infants from the streets and from the view of the community. Parents were able to anonymously drop their infants and children in a wheel that turned into a building that accepted these children. In France, these wheels were called les tours. The state and church organizations took responsibility for these children, enabling the parents, relatives, and society to forget about them (Boswell, 1988). The foundling system grew into an important and necessary institution. By the late eighteenth century in Toulouse, France, 25 percent of children were known to be abandoned; in poor areas the rate was nearly 40 percent, while in rich areas that abandonment rate was 15 percent. By 1784 in all of France, there were 40,000 abandoned babies, and by 1822 there were 138,000 babies abandoned. Between 1827 and 1833, over 336,000 babies were left in foundling homes (Schwartz & Isser, 2000). The historical record in the nineteenth century France showed that single women abandoned their infants at twice the rate of married women. Between 1830 and 1860, women between 20 and 29 years had a greater tendency to abandon their infants than either younger or older women. For example, single domestic servants abandoned their infants at a higher rate than laundresses or seamstresses because their single status along with the demands of their jobs made caring for a child nearly impossible (Fuchs, 1987). Unfortunately, the foundling home system was ill-equipped to deal with the large numbers of children who were abandoned. The shortage of wet nurses, overcrowding, lack of sanitation, and diseases contributed to the high-mortality rates, which averaged 60 percent in France in the eighteenth century (Fildes, 1988). States in Europe tried to restrict abandonment to just
The States’ Response to Parental Divestment
17
illegitimate children, but it was very difficult to enforce this rule as even married couples were abandoning their infants. In some places in Europe, the mortality rate reached 90 percent as most children died either from communicable disease or from a lack of nutrition because their care depended on over-extended wet nurses. Most people probably had no idea about the atrocious conditions or the high-death rate in the foundling homes (Boswell, 1988; Fildes, 1988; Sa, 2000). Nonetheless, according to Fildes (1988), ‘‘y these institutions, set up in the light of Christian charity and often run by genuinely-concerned administrators, were an alternative method of infanticide for desperate mothers. At least if the child was abandoned there was a small chance it would survive’’ (p. 158). There are several possible reasons why some eighteenth- and nineteenthcentury parents decided to abandon their infants. Some parents were unable to financially support these infants because of poverty or disaster. Others abandoned their infants because of the shame of illegitimacy or incest. In France during the nineteenth century, two-thirds of abandoned children were the result of illegitimate relationships (Fuchs, 1987). While some parents abandoned their infants because they had no interest in parenthood, others did so in hopes for better lives for their children. Other parents felt compelled to leave their infants at foundling homes because they needed to conserve already-scarce resources for an older child. Finally, in Europe some infants were abandoned because they were the wrong gender, that is, female (Boswell, 1988). For centuries in Europe, abandoning parents faced no serious sanctions from the state or society. While abandonment was widely practiced, there were groups that criticized it. Ancient and Christian writers condemned abandonment because it was the result of irresponsible sexuality, and because there was a possibility of incest if a child grew up to have future sexual relations with an unknown relative (Boswell, 1988). Other critics considered infant abandonment to be a dereliction of duty to the family and even the state. In some countries in Europe, parents were able to reclaim their children at a later time, thus leaving the child to be cared for at the public’s expense. Within the foundling system, sometimes a mother would arrange to become her own child’s wet nurse, thus getting paid to feed her own child (Sa, 2000). Finally, it is interesting to note that critics of infant abandonment make no mention of an inherent obligation of a parent to his or her child (Boswell, 1988). These foundling systems in Europe lasted for centuries because they not only served the needs of parents, but also that of the state. According to Sa (2000), foundling children who survived were adopted and served the needs
18
LAURETTE T. LIESEN
of the state. They were used as laborers or were enrolled in the army. Others were sent overseas to colonize new territories for European countries. By the mid-nineteenth century with the foundling systems becoming stretched beyond capacity, European nations passed laws restricting the abandonment of children, and established a new welfare system for mothers, especially for those who were unwed. For example, France passed laws that restricted child abandonment and replaced foundling homes with aid to unwed mothers as a means to prevent abandonment. Part of the aid to unwed mothers who could not afford a mid-wife was the establishment of maternite hospitals. They admitted only those women who agreed to nurse their infants for several days while in the hospital, and they received aid only if they promised not to abandon their children. This system did successfully reduce the number of children sent to the foundling homes. In 1830, 64 percent of children were sent to foundling homes, and by 1838 around 40 percent were left (Fuchs, 1987). By 1852 the women who went to maternite hospitals in France had to stay eight days, nurse their infants, and take them home with them. As a result of these new rules in France, fewer women stayed at the hospital. Nonetheless, among those who did, by 1855 only 5 percent abandoned their babies in these maternite hospitals because many women had the time to establish bonds with their infants (Fuchs, 1987). Even though the maternite hospitals reduced the number of infant abandonments shortly after birth, many women left these hospitals only to abandon their children to the foundling homes within a month. It appeared that the amount of state aid given to these women did impact their decisions not to abandon shortly after their infants were born. Initially, the aid amounts given to women were small and only a few women received aid. In the 1830s, 2 percent of women received state aid, and by 1869 31 percent of women received it. As the state aid increased, infant and child abandonment decreased in France. This was especially evident among poor, married women (Fuchs, 1987). While child abandonment did not stop completely, the dismantling of the foundling system greatly reduced the easy and convenient opportunities for parents to divest from their children.
CONTEMPORARY ANALYSES OF CHILD ABANDONMENT In the United States in the late 1990s, several researchers attempted to explain the series of high-profile cases of young women abandoning or killing their
The States’ Response to Parental Divestment
19
infants in the United States. Rather than turning to the evolutionary sciences for insight, they used psychological and sociological explanations in their analyses. For example, Schwartz and Isser (2000) make the distinction between infant and child abandonment. If an infant is killed at or within hours of birth, even when abandoned, it is considered neonaticide. If a child is murdered within the first year of life, it is considered infanticide. They found that neonaticide is usually committed by a mother who gives birth unattended, while infanticide is committed by one or both parents. In another study on neonaticide, Meyer and Oberman (2001) show that many women who abandon their infants have a specific set of psychological and sociological characteristics. They tend to be young, single women who did not disclose their pregnancies to family members or friends. The fathers of these abandoned infants are not involved in these mothers’ lives. These women gave birth alone and in secret. According to Meyer and Oberman’s (2001) study, these women experienced intense shame, guilt, and fear about their pregnancies. The young women they interviewed were so fearful of disappointing their parents or others in their communities that they were in denial of their pregnancies. Amazingly, these women lived with their families and interacted with others who were oblivious to these young women’s physical changes. These young women never confided to anyone about their pregnancies. In another more in-depth study, Oberman and Meyer (2008) interviewed women in prison who killed their infants. They all had similar experiences of isolation from family and friends (even if they lived with or near family members), the violence of rapes or beatings, and an overall sense of hopelessness. In another analysis of infant abandonment and current safe haven laws, Sanger (2006) links the popularity of this legislation in the United States with the moral concern about abortion and the emergence of a ‘‘culture of life,’’ which aims to protect all human life. She maintains that safe haven laws portray ‘‘mothers as untrustworthy persons by reinforcing the proposition that women who abort and mothers who abandon newborns are the same: Both kill babies’’ (Sanger, 2006, p. 761). Because many young women have absorbed the culture of life message that sex before marriage and abortion is wrong, a small number of young women become morally paralyzed by denying their pregnancies and then abandoning their infants. According to Sanger (2006), the fetus has become an important player in American social and political life, thus contributing to the increase in infant abandonment. While these analyses are helpful in understanding the type of women who are at risk of abandoning their newborns, they are incomplete. They do not
20
LAURETTE T. LIESEN
explain why women choose to divest from these infants. An evolutionary perspective explains why child abandonment and infanticide persist among some human communities and why it may be considered beneficial reproductive strategy in the minds of these mothers in terms of their longterm reproductive success.
EVOLUTIONARY PERSPECTIVES ON CHILD ABANDONMENT AND INFANTICIDE As Sarah Blaffer Hrdy (1999) argued in her book Mother Nature, conflicts of interest arise between a mother and her child. Mothers at times are forced to make trade offs during their reproductive years – produce many offspring and invest minimally in each, or produce a few offspring and invest a great deal in those few. According to Trivers (1972), parental investment is any investment by a parent in an individual offspring that increases the offspring’s chance of surviving at the cost of the parent’s ability to invest in additional offspring. This investment includes the metabolic investment in producing the primary sex cells as well as any investment that benefits the offspring, such as feeding or protecting them. Each offspring can be considered an investment independent of other offspring; increasing the investment in one offspring tends to decrease the investment in another. Therefore, a large parental investment is one that reduces the parent’s ability to produce another offspring. Among mammals generally, the female’s parental investment in offspring is considerable, especially because of the lengthy period devoted to gestation and lactation, thus limiting her capacity to produce another offspring. Primate mothers, especially women, invest a disproportionate amount of time, energy, and physical resources in gestating, lactating, and caring for their offspring for many years. As Hrdy (1999) explained, mothers must compromise and balance their own subsistence needs, time, energy, and resources they need to survive, mate, and reproduce. Depending on the environment and circumstances, mothers may choose to terminate their care of an infant when it is in their reproductive interests to do so. It is important to note that human mothers do make fitness trade-offs, and these decisions are heavily influenced by a mother’s social network – mates, future mates, biological relatives, and other members of her community. For example, in South America, Ayoreo, a mother who lost her infant’s father, tended to opt for infanticide since the child’s long-term survival depended on its father’s protection and resources (Hrdy, 1992).
The States’ Response to Parental Divestment
21
Termination of parental investment can range from the act of committing infanticide or the more passive act of abandonment. While there are legal and moral differences between the two types of disinvestment, from a biological perspective, the two are similar (Hrdy, 1999). Throughout human evolutionary history, some mothers killed their infants as a type of birth control. According to Hrdy (1992), biological parents are responsible for the largest number of infanticides. Marriage, inheritance systems, religious beliefs, and social norms concerning individual and family honor play a central role in parental decisions to abandon a child or commit infanticide. In Hausfater and Hrdy’s (1984) anthology that examined infanticide across species and from an evolutionary perspective, they define infanticide as ‘‘any form of lethal curtailment of parental investment in an offspring’’ (p. xv). This killing of offspring is done by conspecifics in a variety of ways – destroyed gametes, reabsorbed fetus, abortion, or killing an infant or child. Rather than seeing infanticide as pathological (or immoral behavior among humans), evolutionists see infanticide as functionally adaptive with reproductive benefits for the individual committing the infanticide. The victims of infanticide are vulnerable, and their survival will lead to significant future expenditures of costly resources for the parents. According to Hausfater and Hrdy (1984), there are five functional categories of infanticide across species: Exploitation of the infant as a resource, such as cannibalism. Competition for resources where the death of an infant increases the availability of resources to the killer or the family as a whole. Sexual selection where one sex is killed in preference of the other sex, and it is usually a female child that is killed. Social pathology where illness leads a parent to kill its offspring. Parental manipulation of progeny in which the parents increase their own lifetime reproductive success. The last category of parental manipulation is prevalent among humans. This is a situation in which the death of the infant will sometimes improve the chances of survival for the mother, an older offspring, or lead to greater net reproductive fitness for the mother, father, or both. Infanticide has been used in human populations to space births, enabling the mother to raise more children to adulthood, thus maximizing her individual reproductive success (Scrimshaw, 1984). If an infant is the wrong sex, illegitimate, or deformed, some traditional cultures will leave an infant to die. In a more recent study, Beaulieu and Bugental (2008) found that mothers who suffer from deep depression have a contingent relationship with their infants if
22
LAURETTE T. LIESEN
they were born prematurely, thus having low reproductive potential. They suggest that these infants with these types of mothers are at risk of neglect, abuse, or possibly infanticide. Hrdy (1992) states that mothers with highreproductive values, that is, many more years ahead for reproducing, are significantly more likely to divest their interests in a child born too soon. Finally, poor ecological conditions causing a shortage of resources may lead parents to commit infanticide (Hausfater & Hrdy, 1984; Scrimshaw, 1984; Hrdy, 1992). However, it is important to note that poverty itself is a poor predictor of infant abandonment and infanticide. Kin assistance and family organization have been very important factors in communities where there is little child abandonment (Hrdy, 1999). According to Scrimshaw (1984), the infanticide committed by humans is a type of behavior that ranges from the deliberate to the unconscious. Infanticide includes the active killing of an infant, placing the infant in dangerous situations, abandonment where survival may be possible, ‘‘accidents,’’ excessive physical punishment, lowered biological support, and lowered emotional support. While Scrimshaw concurs with Hausfater and Hrdy (1984) that infanticide is used by individuals and within families to control fertility, she also points out that there are societies, such as the ancient Greeks, the Japanese, the Chinese, and Eskimo communities, that explicitly practice infanticide to limit their populations. Accurate data on infanticide across cultures is nearly impossible to gather. Nonetheless, anthropologists in the Human Relations Area Files show that infanticide has been practiced for a long time by human societies. Daly and Wilson (1984) state that 36 percent of preindustrial cultures practiced infanticide and 13 percent of cultures practiced it occasionally. They report that nonpaternity, deformity, and inadequate parental resources were among the reasons given for infanticide. In terms of parental resources, it was the mothers who felt that they did not have the capacity to raise the infant because they were overburdened, they had twins, the timing was poor, or they were unwed. If the mother died in childbirth, some cultures also killed the baby since there was no one to care for it (Daly & Wilson, 1984). A Western analysis of infanticide was conducted by Daly and Wilson (1984). Based on an examination of Canadian homicide statistics from 1961 to 1979, they found patterns of risk among women for infanticide. While they could not attribute motives, such as dubious paternity or deformity, they did find that the probability of child homicide committed by a parent is highest when a child is an infant, and then rapidly declines with age. They also found that the mothers who committed infanticide were often single.
The States’ Response to Parental Divestment
23
Between 1977 and 1979, 60 percent of infanticidal mothers were unwed. In addition, mothers who committed infanticide tended to be young – over 15 percent were 17 years old or younger compared to 3 percent of the population who were new mothers. In the United States, the statistical patterns of infanticide are similar to Daly and Wilson’s (1984) findings in Canada. According to the U.S. Bureau of Justice Statistics (U.S. Department of Justice, 2007), of the total number of children killed under the age of four years, most are killed before they reach one year of age. For example, in 2003 there were 610 infanticides of children up to four years old. Two hundred and fifty-eight children were under one year when they were killed. Similar to the Canadian statistics, in the United States the chances of children being killed declines with each year of life. For 2003 over 56 percent of the offenders of child homicides were a parent of that child victim. In the West, abandonment has been practiced more often than active infanticide. In the United States most infants are abandoned at hospitals shortly after delivery (Appell, 2002a). According to a study by the U.S. Department of Health and Human Services (1994), most abandoned infants were left in hospitals by the parents who were unable to care for them. The study found that during 1991 there were 12,000 abandoned infants in U.S. hospitals. Over three-fourths were exposed to drugs while in utero, born prematurely, and had low-birth weight. Fifty-seven percent of the abandoned infants had medical complications, such as respiratory problems, fetal alcohol syndrome, feeding disorders, mental retardation, genetic disorders, cleft palates, and other anomolies. By 1998 there were 30,800 children abandoned in U.S. hospitals, a 43 percent increase. Overall, an evolutionary perspective on child abandonment and infanticide not only supports the historical, psychological, and sociological analyses, it also provides a framework for understanding why parents throughout human history and across cultures under certain environmental conditions and personal circumstances have chosen to divest from their children. As parental investment theory recognizes, each child demands intense and lengthy investment of personal energy and resources to care for her even when she is healthy and all the necessary emotional, social, and financial resources are available. If a parent (particularly the mother) cannot make this commitment, she may try to divest from that child through abandonment or infanticide. The crime statistics show that mothers who commit infanticide are emotionally, socially, or financially ill-prepared to be mothers. Those infants that are abandoned in hospitals tend to demand an inordinate amount of care and resources because of their physical problems,
24
LAURETTE T. LIESEN
which are probably why the mothers abandoned them there. These decisions are reproductive strategies and are made in terms of mother’s individual survival and long-term reproductive success, which explains why this behavior has persisted throughout human evolution.
CONCLUSION Even though safe haven laws attempt to prevent the deaths of abandoned infants and appear to be a compassionte response to parents who are in distress, these laws may inadvertently lead to more parents abandoning their children. There are several problems with the contemporary safe haven laws that may lead to an increase in child abandonment. First, many state laws today are written too broadly in that these laws apply to both newborns and children up to one year of age. This provides an extended window of opportunity for parents to divest from their children when it becomes in their interest to do so. This was evident for centuries in Europe with foundling homes that took infants at anytime and anonymously, and in Nebraska in 2008 with its open-ended age limit for abandoning children. According to Appell (2002a), safe haven laws are also written too narrowly in that they in no way address the factors that lead mothers and parents to kill or abandon their infants. As both criminal statistics and evolutionary perspectives indicate, infants are most at risk in the hours after birth. Safe haven laws do not address the young, single women who are in denial of their pregnancies and are at the greatest risk of killing or abandoning their children. If an evolutionary perspective were considered in the development of this legislation, states may have expanded policies and programs that would better support young pregnant women via social networks, access to free prenatal care, and financial assistance. Safe haven laws are wide spread because they are ‘‘feel good’’ pieces of legislation that cost little to nothing and allows politicians to be perceived as being compassionate and caring for babies. Yet, safe haven laws abandon the mothers who are struggling, and fail to address the factors that lead women to abandon their infants. Even though there is little data currently collected by the states, historical and cross-cultural analyses, and explanations from evolutionary theory support the argument that safe haven laws will probably increase infant and child abandonment. Instead of seeing infant abandonment as a new phenomena, this analysis sees infant abandonment as part of human evolutionary history and one of several reproductive strategies individuals will use to insure their overall reproductive success. Parents, especially
The States’ Response to Parental Divestment
25
mothers, have always assessed their ability to successfully raise their children. Some parents decide to divest in order to improve their overall reproductive success. These choices are evident across cultures and throughout Western history. The French experience with its foundling system led to an increase in infant abandonment instead of a decrease. Many mothers, not just the poor, took advantage of this system to disinvest from their children. In contemporary Western societies, most decisions of noninvestment occur via contraceptive use, and most decisions of divestment occur well before an infant is born via abortion. The women who divest from their infants after birth are making similar decisions, albeit delayed and if not conflicted, about whether to raise the child or not.
ACKNOWLEDGMENTS I would like to thank both Lewis University for its financial support for this research project, and Dr. Robert Sprinkle for his comments on an earlier draft. Finally, I extend my deepest appreciation to my husband David and my children Matthew and Carolyn for their support while I worked on this project.
REFERENCES Appell, A. R. (2002a). Safe havens to abandon babies, Part II: The fit. Adoption Quarterly, 6(1), 61–69. Appell, A. R. (2002b). Safe havens to abandon babies, Part I: The law. Adoption Quarterly, 5(4), 59–69. Baran, A. (2003). Unintended consequences: ‘Safe Haven’ laws are causing problems, not solving them. New York: Evan B. Donaldson Adoption Institution. Available at http:// www.adoptioninstitute.org. Last viewed on October 4, 2010. Beaulieu, D., & Bugental, D. (2008). Contingent parental investment: An evolutionary framework for understanding early interaction between mothers and children. Evolution and Human Behavior, 29, 249–255. Boswell, J. (1988). The kindness of strangers: The abandonment of children in Western Europe from late antiquity to the renaissance. New York: Pantheon Books. Daly, M., & Wilson, M. (1984). A sociobiological analysis of human infanticide. In: G. Hausfater & S. Hrdy (Eds), Infanticide: Comparative and evolutionary perspectives (pp. 487–502). New York: Aldine. Eckholm, E. (2008a). Special session called on Nebraska safe-haven law. New York Times, 30 October, p. 29. Eckholm, E. (2008b). Nebraska limits safe-haven law to infants. New York Times, 30 November, p. 10.
26
LAURETTE T. LIESEN
Fifty-three safe babies – So far. (2009). Editorial, Chicago Tribune, 17 August, p. 14. Fildes, V. (1988). Wet-nursing: A history from antiquity to the present. New York: Basil Blackwell, Inc. Fuchs, R. G. (1987). Legislation, poverty, and child abandonment in nineteenth century Paris. Journal of Interdisciplinary History, 18(1), 55–80. Greimel, H. (2007). Abandoned toddler spurs debate in Japan. Chicago Tribune, 17 May, p. 10. Guttmacher Institute. (2010). State policies in brief: Infant abandonment. 1 October. Available at www.guttmacher.org. Last viewed on October 6, 2010. Hausfater, G., & Hrdy, S. (1984). Comparative and evolutionary perspectives on infanticide: Introduction and overview. In: G. Hausfater & S. Hrdy (Eds), Infanticide: Comparative and evolutionary perspectives (pp. xii–xx). New York: Aldine. Hrdy, S. (1992). Fitness tradeoffs in the history and evolution of delegated mothering with special reference to wet-nursing, abandonment and infanticide. Ethology and Sociobiology, 13, 409–442. Hrdy, S. (1999). Mother nature: A history of mothers, infants and natural selection. New York: Pantheon Books. Lawmakers vote to let mothers leave newborns in safe places. (2001). 22 May, Associated Press State & Local Wire. Available at www.lexis-nexis.com/universe/docum. Last viewed on June 15, 2001. Marrazzo, A. (2009). Revised law has 30-day window. Chicago Tribune, 25 December, pp. C-1, 3. Meyer, C., & Oberman, M. (2001). Mothers who kill their children: Understanding the acts of moms from Susan Smith to the ‘‘Prom Mom’’. New York: NYU Press. Oberman, M., & Meyer, C. (2008). When mothers kill: Interviews from prison. New York: NYU Press. Povoledo, E. (2007). Updating an old way to leave the baby on the doorstep. New York Times, 28 February, p. A4. Ross, T. (2008). Nebraska dad who left 9 kids says he was overwhelmed. Yahoo News, 26 September. Available at http://news.yahoo.com/s/ap/s0080926/ap_re_us/ children_safe_haven. Last viewed on September 26, 2008. Sa, I. (2000). Circulation of children in Portugal. In: C. Panter-Brick (Ed.), Abandoned children (pp. 27–40). Cambridge: Cambridge University Press. Safe Place for Newborns: About the Program: Facts and Stats. (2001). Safe place for newborns. Safe Place for Newborns: About the Program: Facts and Stats, 21 March. Available at http://www.safeplacefornewborns.org/factsnstats.html. Last viewed on July 17, 2001. Sanger, C. (2006). Infant safe haven laws: Legislating in the culture of life. Columbia Law Review, 106(4), 753–829. Save Abandoned Babies Foundation. (2009). Timeline. Save Abondonded Babies Foundation, August 12. Available at http://www.saveabandonedbabies.org. Last viewed on August 22, 2009. Save Abandoned Babies Foundation. (2010). Illinois safe haven law improved. July 20. Available at http://www.saveabandonedbabies.org. Last viewed on October 4, 2010. Schwartz, L., & Isser, L. (2000). Endangered children: Neonaticide, infanticide, and filicide. Boca Raton, FL: CRC Press. Scrimshaw, S. C. (1984). Infanticide in human populations: Societal and individual concerns. In: G. Hausfater & S. Hrdy (Eds), Infanticide: Comparative and evolutionary perspectives (pp. 440–461). New York: Aldine.
The States’ Response to Parental Divestment
27
Slevin, P. (2008). Nebraska to alter safe-haven law: State hopes to care for abandoned children without becoming a dumping ground. Washington Post, November 16, p. A03. Trivers, R. (1972). Parental investment and sexual selection. In: E. Campbell (Ed.), Sexual selection and the descent of man, 1871–1971 (pp. 136–179). Chicago: Aldine Publishing. U.S. Department of Health and Human Services. (1994). Report to congress: National estimates on the number of boarder babies, the cost of their care, and the number of abandoned infants. U.S. Government Printing Office, Washington, DC. U.S. Department of Justice. (2007). Bureau of justice statistics homicide trends in the United States: Trends in infanticide, July. Available at http://www.ojp.usdoj.gov/homicide/ tables/kidsagetab.htm. Last viewed on August 3, 2009.
A MODEL OF ACTIONS AND NORMS: AN INTEGRATED EVOLUTIONARY PERSPECTIVE ON NORMATIVE ETHICS AND HUMAN BEHAVIOR Birgitta S. Tullberg and Jan Tullberg ABSTRACT One fundamental question in normative ethics concerns how norms influence human behavior and discussions within normative ethics would be facilitated by a classification that treats human actions/behavior and moral norms within the same functional framework. Based on evolutionary analysis of benefits and costs, we distinguish five categories of human action. Four of these – self-interest, kin selection, group egoism, and cooperation – are basically results of gene selection, benefit the individual’s genetic interest and may be described as ‘‘broad self-interest.’’ In contrast, the fifth category, unselfishness, is more likely a result of cultural influences. All the five categories of action are influenced by three broad moral spheres, each of which represents many norms that have a common denominator. Thus, a sphere of integrity concerns the individual’s right to act in his/her interest and against those of other individuals. A sphere of reciprocal morality deals with rules for various Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 29–51 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009004
29
30
BIRGITTA S. TULLBERG AND JAN TULLBERG
forms of cooperation. An altruistic sphere has to do with the obligations to generate advantages for others. Ethics can be viewed as a dynamic conflict among various norms within and between these spheres. The classical conflict is that between the integrity and altruistic spheres. However, we argue that the prime antagonism may be that between the altruistic and reciprocal spheres; the main impact of altruistic ideals may not be the reputed one of counteracting egoism, but subversively thwarting reciprocal morality.
INTRODUCTION Most of us hold strong moral opinions and make daily judgments of right and wrong. At the same time, academic moral philosophy is often regarded as an abstract topic without practical relevance. Yet, most people do believe that ethical norms have a great influence on the morals we practice. Even if we do not obey all commandments, but reserve individual freedom of judgment, we are directly or indirectly affected by these rules; our exemption does not eliminate all effects of the general principles. The choice of moral principles is therefore most important, something that concerns each and everyone, and not something to be left for ethical committees or philosophers. This chapter is divided into six parts: 1. The relationship between evolution and ethics is briefly discussed in the introduction. 2. Human actions are analyzed in an evolutionary framework. 3. A broad classification of ethical rules is suggested. 4. Possible causes of human altruism are discussed. 5. Moral conflicts are discussed in terms of competition between spheres. 6. Conclusions are drawn about how different classes of ethical rules may affect human behavior and sociality. Much of human behavior can be understood and explained from an evolutionary perspective. The fact that the same evolutionary processes have shaped all life forms, and that we share features with other organisms through a common evolutionary history speak for incorporating human behavior in a Darwinian scheme. Still, morality is often regarded as a unique human feature, and this may be one reason for even many biologists being
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
31
reluctant to discuss moral philosophy from an evolutionary perspective. Huxley (1894), for instance, regarded ethics, the ‘‘moral process,’’ as antagonistic to evolution, the ‘‘cosmic process,’’ and Williams (1989) follows Huxley’s view that nature is an enemy that has to be combated. Nevertheless, over the last couple of decades, there has been an increased interest in the interrelationship between evolutionary biology and moral philosophy (Alexander, 1987; Campbell, 1975; de Waal, 1996; Frank, 1988; Masters, 1983; Ridley, 1996; Ruse, 1986; Wilson, 1975, 1979; Wright, 1994). For instance, based on studies on monkeys and apes that show phenomena like the capacity for empathy/sympathy, mutual aid and conflict resolution, de Waal (1996) concludes that ‘‘evolution has produced the requisites for morality,’’ and therefore, that there is no fundamental conflict between evolution and ethics. Because of the impact of evolution on human behavior, it is a factor that needs to be taken into account in a discussion about ethics. An evolutionary perspective can shed light on and explain a range of human actions. First, our psychological constitution, which is a result of an evolutionary process, has a direct effect on our actions (e.g., Barkow, Cosmides, & Tooby, 1992; Pinker, 2002). At the same time, our psychological functioning is affected by the exposure to specific norms, which in turn are a product of the human psyche (e.g., Richerson, Boyd, & Henrich, 2003). Thus, because norms are shaped by humans, our evolutionary history is likely to indirectly affect the design of those norms. Yet, moral philosophy is difficult, for several reasons (e.g., Wilson, 1998), and we agree with the philosopher Elliott Sober in that ‘‘It is not implausible to think that many of our current ethical beliefs are confused. I am inclined to think that morality is one of the last frontiers that human knowledge can aspire to cross’’ (Sober, 1993, p. 208). According to Sober, one reason why the question about how we ought to lead our life is so difficult to come to grips with is because it is clouded by self-deception (also see Trivers, 1985). Because of the natural relationship between moral norms and human actions (via the human psyche), it would be valuable to be able to discuss these within the same general framework. One framework, commonly used in evolutionary biology, is to consider the general effects of behavior in terms of fitness, or correlates to fitness. Such effects are commonly measured as gains and losses in materialistic, survival, and, ultimately, reproductive, terms. In our view, ethical rules too can be considered within such a framework, that is, who will gain/lose by others following a rule, and who will gain/lose by following a rule. Such a functional perspective on ethics may shed light on potential conflicts as well as agreements among various norms. We believe
32
BIRGITTA S. TULLBERG AND JAN TULLBERG
that a treatment of actions and norms within this general framework would facilitate analysis within normative ethics, and the purpose with this chapter is to contribute to analytical progress within this field. It is important to note that although behavior can be seen as given in an evolutionary perspective, in normative ethics the impact of norms on behavior is clearly acknowledged and in focus. Accordingly, in normative ethics behavior can be regarded as the dependent variable and ethics as the independent variable, and the fundamental question concerns how ethical norms will influence behavior.
ACTIONS IN AN EVOLUTIONARY FRAMEWORK On the basis of functional design and effects, in terms of costs and benefits to agents, we identify five broad categories of action. These are self-interest, kin selection, cooperation, group egoism, and unselfishness, and we discuss each of these categories in turn. It is important to note that the categories are primarily descriptive, not normative, and that each category contains both what we would consider normatively good and bad actions. Various classifications of human behavior have been proposed previously (Alexander, 1987; Campbell, 1975; Masters, 1983, 1989; van den Berghe, 1981) and we discuss some of these below. Actions could potentially be categorized on several grounds: (1) by the actor’s intention and (2) by actual effect or consequence. However, we use a third ground for categorization of actions, namely (3) the general effect of actions with a certain design. Here, the action is not regarded as an isolated event but as a part of a pattern or strategy. The result may vary, but it is the statistically expected value that is most interesting when understanding the existence of a behavior. Since ‘‘expected’’ carry a connotation implying intention, we use the term ‘‘likely effect.’’ For instance, an investment of money that happens to be unsuccessful is still classified as a self-interested action. In general, we would expect a high degree of correlation between intention, actual effect, and likely effect of an action. In moral philosophy and law, intentions are judged extremely important (cf. man-slaughter vs. murder). Intention, or purpose, can be estimated from either an agent’s own claim or a bystander’s observation. The fact that these do not always agree shows that there are severe empirical problems with an intentional approach. Moreover, we could think of certain actions, for example, saving one’s child from drowning, being carried out in an almost reflexive or automatic manner, without being preceded by a calculated decision. The
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
33
problem with a classification based on intentions would be where to place such actions. In contrast, no conscious premeditation of an action is needed in order to classify it based on likely effects. Neither does inadequate intelligence nor self-deception, influence such a classification. Subjective beliefs or claims of good intentions can be important parts of the behavioral mechanisms, but they are not decisive for our more ultimate classification. We also leave out emotions from our basic classification and this point needs elaboration. We regard emotions as explanations for behavior on a proximate level, and we also acknowledge that these are the targets of behavioral evolution. There are specific emotions connected to each of our behavioral categories, and also many different emotions behind each category. For instance, a reciprocal behavior may be triggered both through a positive feeling of gratitude or a more negative feeling of guilt or revenge. The ultimate reason for the existence of such feelings may be that they tend to enforce behavior toward a larger long-term benefit (through cooperation) at the expense of a smaller short-term gain (through deceit) (Frank, 1988). Although the emotional background of behavior is important in its own right, there is a tendency to give emotions an overall importance when classifying actions. For instance, it has been argued that real altruism does not exist because, for example, saving a total stranger in distress either relieves the distress felt by the actor himself or generates a feeling of goodness in the actor. Both kinds of emotional change are for the better to the actor, and should be regarded as basically selfish. Altruism is thus defined away. However, here we define the act as altruistic because of its likely effect in materialistic terms. In conclusion then, psychological factors related to intention, emotion, and cognition are decisive for the execution of various actions. However, here we regard them as intermediate factors and base our classification of actions on likely effects.
Self-Interest Selfishness is hard to define in a manner both clear and consistent with ordinary usage. One problem with selfishness – and even more with its synonym egoism – is that, in a moral sense, it gives rise to associations with many things we dislike. Moreover, there are other actions that we gladly perform but do not enjoy calling egoistic, but maybe self-realization. Therefore, we have chosen the term self-interest to encompass actions with an expected benefit for the actor without regard to how others are affected.
34
BIRGITTA S. TULLBERG AND JAN TULLBERG
It can always be argued that we could direct time and effort to the benefit of others rather than to ourselves so that others suffer an alternative loss when we favor our own interests. A functional definition of self-interest is as follows: ‘‘actions that on average produce greater advantages than disadvantages for the actor.’’ The definition includes all actions bringing an expected net benefit to the actor independent of whether they are considered self-evidently justified or morally condemnable. On these grounds, self-interest is plainly a label that suits a great deal of what we do, ranging from actions that maintain our physical selves – to the social situations where we behave in our own interest. Most such actions are, in fact, so automatic that we give them no reflection at all. In regard to numerous other actions, we think it so obvious that our personal preferences should lead the way, that we see no moral choice confronting us. Actions for which a self-interested option is questioned are a tiny portion of the total.
Kin Selection Kin selection is a notably important process in understanding the evolution of behavior in social animals, not least people (Hamilton, 1964). Whichever human society is studied, helping own kin is common. For example, the substantial sacrifices we make for our children are a very large proportion of what we do for others. The function is sometimes said to be mutual: we help our children so as to be helped in the future. Such alliances do exist in various cultures, but are not the decisive ones. A child’s future sacrifices are primarily devoted to its own children, not its parents. We are ready to take a conscious loss in dealing with our children because of genetic, rather than individual, rationality. Many actions in this category, such as helping children and other relatives, are generally considered good or acceptable. However, as with each of the categories this one contains actions that may be disproved of, and nepotism is a word of normally negative import. Even cultures that are critical of nepotism in principle will exhibit systematic use of it.
Cooperation Cooperation among unrelated individuals is important in all human societies. David Hume (1739) spoke of ‘‘confined generosity’’ and a synonymous label is ‘‘reciprocal altruism’’(Trivers, 1971; Humphrey, 1997).
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
35
The importance of ‘‘contacts,’’ a dynamic network of services and return services, is emphasized in all areas of social life. If reciprocal behavior is to develop, someone must take the first step by making a sacrifice that benefits others. This does not always succeed, of course; nor does it function perfectly even in a group with few individuals. But there is often a chance to perform a deed, which is a minor sacrifice, but of greater value to the recipient, which, in turn, motivates a return service that benefits the first giver. Such an increase in effect means that reciprocity can survive in spite of the waste that occurs when some services are never returned. A classical example of reciprocal behavior in animals is the study by Wilkinson (1984) on vampire bats. A recent example concerns joint mobbing of predators by flycatchers, where birds help others from which they have previously received help, but refrain from supporting non-helpers or cheaters (Krams, Krama, & Igaune, 2006). Central to reciprocity are return services (Axelrod, 1984), and a capacity to behave reciprocally has been anchored in many emotive responses (Alexander, 1989; Fessler & Haley, 2003). For instance, sympathy is mutual to a striking extent. The debt of gratitude we feel on neglecting our part of a reciprocal relationship is an emotional reinforcement of behavior, which has demonstrated its strength in the process of evolution. The ability to detect cheaters is also likely to have evolved in a reciprocal context (Cosmides, 1989). As in all the categories of action, both a good and a bad side exist here, quite close together. Friendship is regarded as good, while the same action may be called ‘‘friendship corruption’’ or ‘‘partial behavior’’ by someone who has suffered from others’ collaboration.
Group Egoism Yet another type of behavior can be called ‘‘accumulative egoism.’’ In sociobiology, the phenomenon may be described as ‘‘gregariousness’’ or ‘‘selfish herd’’ (Hamilton, 1971). Here, the term ‘‘group egoism’’ will be employed. It is essential to recognize individual interest as the central point of departure, the group being primarily an instrument. Thus, group egoism builds upon individual rationality. By joining forces, individuals increase the possibility of improving their conditions. Human society involves many actions and institutions that can be attributed mainly to group egoism. It comprises much of the activity in human society, from small gangs up to nations, and occurs spontaneously due to palpable advantages. Labor unions and business enterprises are clear instances.
36
BIRGITTA S. TULLBERG AND JAN TULLBERG
Experiments within the field of social psychology show that there are several powerful psychological mechanisms involved in group-egoistic actions (e.g., Brown, 1986; Cialdini, 1988). Among these are the tendencies to conform to group values, to upgrade members of the own group, and to think in terms of in-group/out-group (Fessler & Haley, 2003; Tullberg & Tullberg, 1997). It is important to draw a clear line between categories of action. Cooperation and group egoism are particularly easy to confuse. If, for example, an action benefits people with the same school tie or the same profession, this is not cooperation but group egoism. One can expect members who see one obeying the group norm to consider one loyal and offer help; and the group is strengthened if solidarity is not shown toward members who betray the group’s fellowship. For instance, union members often think less of strikebreakers than of anybody else, and many religious groups consider heresy a worse sin than paganism. In sum, belonging to a certain group is not enough: avoiding and punishing violation of its norms is crucial (Fehr & G€achter, 2002; Fehr & Henrich, 2003; Bernhard, Fischbacher, & Fehr, 2006). There is an ongoing debate in evolutionary biology about the possibilities of group selection in combination with unselfish or altruistic traits (Sober & Wilson, 1998) (next section). A discussion of levels of selection is outside the scope of this chapter, but it has been argued in detail elsewhere that group selection is much more likely to work in combination with group-egoistic traits since they tend to homogenize groups and link group members closer to each other (Tullberg, 2003).
Unselfishness Unselfishness alias altruism is defined as acts that are costly for an actor, do not benefit kin and confer little likelihood of reciprocation. Thus, cooperation and kin selection are not included in this category and this distinction is rather easy to make. However, the distinction between unselfishness and group egoism is more difficult. When are soldiers in a war participating in a group-egoistic versus an altruistic venture? There is a confusion around the term altruism, caused by a use of the term in both a broad (including all our categories but self-interest) and a strict sense, and several workers have pointed out the need to make a clear distinction between these phenomena. Thus, Ruse (1989) uses altruism with and without quotation marks to denote altruism in a broad and strict sense,
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
37
respectively. Lopreato (1981) suggests the term ‘‘ascetic altruism’’ and Hardin (1993) ‘‘promiscuous altruism’’ for altruism in the strict sense, and Boyd and Richerson (1991) have used the term ‘‘self-sacrificial cooperation’’ for the same phenomenon. Here we simply use the terms unselfishness or altruistic behavior. This terminology may also be in line with a general use of the term; for instance Swap (1991) concludes that ‘‘naive’’ perceivers define an action as altruistic when it is directed ‘‘to a needy recipient unrelated to the actor.’’ As mentioned earlier, a motive of self-interest is sometimes inferred for altruistic acts. The giver might find it more worthwhile to feel good when giving away a coat than to keep the coat on. Thus, in a subjective sense, the giver makes no real sacrifice. Similarly, helping a suffering person in order to relieve one’s own distress evoked by watching him suffer is to be regarded as selfish. Then, there is no altruistic behavior, as was to be proved. Without denying that a feeling of goodness as well as stress relief can motivate, we leave out such subjective arguments and take the likely materialistic effects as decisive for our categorization. If a type of action on average yields a sacrifice greater than the positive effects for the subject himself, then the action is altruistic. On the private level, some acts are committed that involve vast sacrifices and deliver huge benefits to the recipient. There are people who donate a kidney to another person (the act belongs to a different category if the recipient is a close relative). Trying to rescue a person from drowning at the risk of one’s life is more often used to illustrate altruistic behavior. Whoever is rescued has every reason to feel a debt of gratitude, and the hero can count on public admiration, so he or she reaps a certain reward. If rescuers are able to minimize their own risks, actions may not result in any personal loss. In many situations, however, the expected result is a loss for the actor. Fifty-six persons were rewarded for selfless heroism in USA and Canada in 1977, whereof eight posthumously (Frank, 1988), and this may serve as an example of existing unselfishness. Far more frequent are smaller acts of altruism such as blood donations – yet many such acts can be interpreted otherwise. Tossing a coin into a collection box is altruistic, unless observed by enough people to qualify as a reputation-building signal.
Influence of Time, Culture, and Norms For our purpose in the present chapter, it is sufficient to conclude that behavior in the five categories mentioned are influenced by a combination of
38
BIRGITTA S. TULLBERG AND JAN TULLBERG
human nature molded by our evolutionary history and the ethical norms that we are exposed to through culture. The speed of economic and technical progress does not blur fundamental similarities in people and their behavior across borders of space and time. However, the amount of actions in various categories may vary; specifically, a shift from the kin selection category to cooperation has occurred with the rise of modern industrialized society (e.g., Richerson et al., 2003). Kin selection, group egoism, and cooperation together account for a large class of actions between the poles of self-interest and unselfishness. Sometimes the chief frontier is seen as that between egoism and other actions. However, a broad self-interest combines four of the categories but excludes altruistic behavior. It should again be emphasized that this classification is functional and not normative. The point of departure is that all categories include actions, which can be seen as good or bad according to one’s norms and values. We consider this method far more fruitful, even in an analysis that is to yield normative conclusions, than the conventional procedure of pitting egoism against altruism.
Comparisons with Other Models After taking part of this model, it is reasonable to ask in what respect it differs from other categorizations of behavior. Compared to the influential sociobiological view as represented by, for example, Edward O. Wilson and Richard Alexander, there is an agreement about egoism, kin selection, and cooperation, but because these three categories together (and extensions such as indirect reciprocity) are seen as including most of human behavior, there is little room for altruism. Thus, perceived deviations are viewed as mistakes or self-deception. In the classical sociobiological scheme, altruism is negligible when it comes to real behavior. Pierre van den Berghe (1981) also includes these major three categories, but adds a fourth labeled coercion. This is certainly a kind of behavior in human societies but to us it looks more like a proximate variable; coercion is a means to enforce rules of self-interest, kin selection or cooperation. Roger Masters (1983, 1989) argues for a scheme introduced by Hamilton, where the three main sociobiological categories are joined by altruism and also mutual harm. We have the same kind of criticism of mutual harm as of coercion, and in this case the lack of evolutionary rationality is more
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
39
evident. Mutual harm can be the outcome of human interaction, but it is hard to give it an independent raison d’^etre. Our model has two distinguishing features. First, it seems motivated to break away group-egoistic actions into an own category because they differ in ultimate rationale as well as emotional underpinning. Second, we conclude that altruistic actions belong to a real, rather than hypothetical, category.
MORAL SPHERES: A BROAD CLASSIFICATION OF ETHICAL RULES Moral systems can be structured in many ways, frequently by setting entire systems against each other, as with Christian and Muslim morality. It may be helpful, instead, to follow a functional division in various dimensions, where classification of moral norms will correspond to the functional classification of actions, that is, depart from the benefits and costs that would result from following the norms. Thinking in this way leads us to a broad classification of moral norms into three spheres: A morality of integrity, which regulates individuals’ rights to act in their own interest and on their own judgments. Norms in this category lead to benefits for me as an individual. A reciprocal morality, which supports mutual benefit and harmonious coexistence. Norms in this category facilitate my possibilities to cooperate with others. An altruistic morality, implying obligations for individuals to follow commandments to selflessly serve their fellow men and ideals. Norms in this category lead to losses for myself and benefits for somebody else. These spheres, too, are primarily functional, not normative. Here no implicit evaluation is made of a sphere as good or bad. Nor is it assumed that moral rules are inherently good and that bad ones are to be categorized as immoral or amoral. The three spheres naturally include principles, which some people praise, but others reject. The purpose of a functional division is to facilitate analysis and promote a well-founded normative assessment. We shall now examine each sphere, in turn, with most examples taken from Western societies with which we are most familiar.
40
BIRGITTA S. TULLBERG AND JAN TULLBERG
Integrity The first questions about the individual’s right to act in his or her own interest are: what should it be called and should it really be treated under the heading of ethics? Our term ‘‘morality of integrity’’ emphasizes the essential concept as well as lessens the confusion that has resulted from the long propaganda war on egoism. The most striking thing about morality of integrity is its expansion during the last few centuries. Previously, many states had an aristocracy with rights in relation to the monarchy, while the common citizen retained little integrity in relation to either these or the priestly elite. With the Enlightenment, and the American and French Revolutions, came new ideas. Conventions on human freedom, from Virginia in 1776 to Helsinki in 1975, have underlined the rights of individuals against the state. The main position is that such human rights are an overriding end in themselves. Another view is that they are functionally valuable in promoting a good society: thus one supports individual rights as a means, not as an end. Both of these views approve of individual interests, even when they conflict with official ‘‘social utility.’’ Individual rights presuppose some economic independence, so that people do not live at the mercy of regimes. Economic life must allow real freedom of choice for the individual, both as a producer and as a consumer. At present, there are few that agitate for total state domination of economic life with very narrow leeway for individuality. When connecting morality of integrity with actions, it is certainly self-interested actions that receive support from the norms in this sphere. Even in decisions that affect oneself very greatly, the individual does not have full freedom of action, and this may seem surprising. According to Catholic belief, suicide is a grave sin, and numerous countries forbid it. The absence of a right to die painlessly can also illustrate the limitations on personal integrity. A right to use drugs is another area where strongly individualistic choices are opposed by different values. It is reasonable to view kin selection as a group of actions influenced by integrity morality. We have a right to take care of ourselves and our children to the best of our ability, but rights over children are complicated. While a genetic concurrence exists that prevents basic conflicts of interest, dire threats to the rights of children do occur. In particular cases, deep feelings may arise for and against the parents: a timely intervention by the social authorities, one asks, or a witch-hunt? Most people probably agree with the parents as responsible guardians, and with the social authorities only in situations of extreme abuse; yet exactly where the border runs is a sensitive
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
41
matter. The growing rights of children can be regarded as an expansion in morality of integrity. Women’s more independent role in the Western society belongs to this development as well. It makes a woman, not chiefly a family member but an individual with goals somewhat different from her husband’s. Two parallel changes have taken place in the family: decisions have been democratized and they have been individualized. Paradoxically, however, a connection is seldom made with self-interest and egoism. Rights to act on one’s own, for personal aims, are not gladly given this label. Whereas ‘‘self-realization’’ is by no means a dirty word, egoism is. What should be equally obvious is that there are moral principles that support the individual’s self-interest and cannot simply be excluded from a moral debate.
Reciprocity This heading has the greatest significance if we look at the social rules that directly affect actions. There has clearly been increased emphasis on cooperation between individuals within the human evolutionary lineage. A fully voluntary reciprocity suffers limitations; a right to give up cooperation at any moment may only yield a game of wait-and-see. Among the principal functions of society is to lay down rules for cooperation, as well as to exert moral and legal pressures on people who break the rules (Gauthier, 1986). The function of many social rules is to facilitate cooperation and coexistence. Traffic laws are a good example. Rules of economic life are mainly intended to create certainty about agreements and obligations; punishment for theft is meant to counteract one-sided transactions that involve no return services. Besides all the regulations of public economic life, there are the reciprocal rules of private life. Here, a rule-breaker risks social repression, not prison. If one expects to be invited back, one must first invite in turn. Gratitude and return favors are constant demands on our behavior. Many of these rules are so prosaic that we see them not as moral rules, but as normal behavior. Yet the actions we really perform are, to a very large extent, based on reciprocal moral rules. Often the two parties in a relationship do not have identical status: man and woman, rich and poor, buyer and seller. The purpose of rules is to establish some sort of balance. In various cultures, the parties are – or were – less equal than in the West today, but a reciprocal undercurrent is
42
BIRGITTA S. TULLBERG AND JAN TULLBERG
still detectable. Its legitimacy may seem dubious, as when a peasant works and his master provides protection. Even so, clarity and acceptance are two conditions for avoiding social conflict. With a weak reciprocal system, economic development is stifled and parasitism becomes rampant. The society turns into a kleptocracy. A functioning reciprocal system is fundamental for every human society. That reciprocal morality promotes reciprocal actions is self-evident, but the link between reciprocal morality and group-egoistic actions is more intricate. The first step, a right to organize groups of people, is based primarily on a morality of integrity. Organizational freedom is closely related to other freedoms of choice, and restricts the state’s power to decide what is a good or a bad organization. However, numerous groups are not principally devoted to an internal activity: they assert the members’ interests against other groups and interests. When these special interests collide, great problems of conflict resolution arise. The special interests in a society must be able to find solutions so that groups do not end in total confrontation with each other: they should be willing to seek compromises. Without mutual advantages, there is no basis for cooperation or, sometimes, even for coexistence. A factor, which enhances the possibilities of avoiding conflict between groups, then, is a strong reciprocal morality. In Western states, a cooperative spirit is widespread. The special interests due to group egoism must, through coalitions and compromise, influence other groupings and ultimately find a resolution acceptable to all. Despite their occasionally predominant rattling of weapons, group-egoistic agitations also presuppose a reciprocal morality of compromise.
Altruism The third sphere, altruistic morality, is the officially dominant philosophy in many societies. One of its products is religion, for which self-sacrifice is always far-reaching. This is most obvious in mortification practices, where one is trained to suppress individual desires and feelings in favor of a ‘‘higher’’ calling. Communist morality runs in the same ruts. A subject’s plain duty is to labor in meekness toward the common goals ordained by the leader. Many people listen with a lump in their throats to the old recording of John F. Kennedy’s presidential inauguration speech: ‘‘Ask not what your country can do for you; ask what you can do for your country.’’
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
43
The individual should make a sacrifice for the state, even if the benefit does not match the effort. While a welfare state has several foundations, one of them is an altruistic claim that we are obliged to help our fellow human beings. The state is then the charitable apparatus, which aids underdeveloped countries and misfortunate citizens: our capacity and generosity provide for their needs and incapacity. The other side of these transfers consists of advantages, which an individual can get from the state: child support, retirement pensions, social payments, resettlement funds and cultural grants. Much in the welfare state could be regarded as a reciprocal system where the costs are covered by the benefits that you receive. A danger in the system, however, is that it is not very transparent, the donor is relatively anonymous and everybody hopes that somebody else is footing the bill. This is, indeed, to be hopeful. Fre´de´ric Bastiat (1848) wrote in the middle of the 19th century: ‘‘The state is the great fiction in which all believe they can live at the cost of others.’’ Although most advocates of altruism would emphasize its peaceful aspects, we must come back to the significance of ethics for warlike aspects. The state’s survival has owed to its maintaining a military strength, which keeps neighboring states out – and which, if possible, can subdue them. A military system does not rely on voluntary assent to its incitements, but traditionally exerts an element of force. The fiery cross had a noose of rope dangling from one end, as a reminder that whoever failed to defend the community could look forward to being hung from its trees. When persuading people of something as repulsive as the duty to risk their lives, it is seldom a matter of using either ‘‘the carrot or the stick,’’ and nearly always of using both together, and this is where altruism serves a function. The soldier needs a higher aim to motivate his great risk. Raising people’s readiness to risk their lives for the state is a serious moral and social task, with various justifications. An historic mission, God’s will, a thousand-year Reich, the triumph of the proletariat, the victory of democracy, and a war to end all wars are among the commonest candidates. The soldier does not intend to die for the cause – ‘‘the earth is strewn y with the graves of men who were slain even as they were inclined to slay’’ (Lopreato, 1981, p. 116). Most deaths could be regarded as unintended altruism (or not as altruism at all by people favoring an intentional definition of actions). But there is a connection between norms and behavior – A member of God’s militia trembles at the threat to his physical existence, but fairly succeeds in persuading himself that his sacrifice is a worthy duty. Altruism is a good carrot, which leads the soldier to gaze upon higher values than his own life. Its ability to motivate is perhaps the chief reason why altruistic acts occur in
44
BIRGITTA S. TULLBERG AND JAN TULLBERG
human cultures (see, e.g., Richerson et al., 2003; Gintis, Bowles, Boyd, & Fehr, 2003). Groups with an ethics of willing sacrifice perhaps possessed, in addition to their apparatus of force, an advantage over groups with that apparatus alone. A capacity for ideological crusades could have been a decisive factor in struggles between competing groups.
UNDERSTANDING HUMAN UNSELFISHNESS Formal analyses have repeatedly shown that altruistic behavior cannot exist as a stable evolutionary strategy (e.g., Maynard Smith & Price, 1973; Williams, 1966; Axelrod, 1984; Axelrod & Hamilton, 1981), but only as a brief form of transition from reciprocal to selfish behavior; reciprocity, but not altruism, may be stable against invasion by a selfish strategy. In this sense altruism is inherently self-destructive. How is then the existence of human altruism to be explained? We believe that prerequisites for altruistic behavior are to be found in each of the other four behavioral categories. For instance, some forms of unselfishness, such as expressed through nursing behavior, are mostly to be traced back to mechanisms operating in a kin selection context. Such acts may be due to a signal-receiver system not being very specific, as for instance when a cry, coming from an unknown child, elicits our empathy and help. In other instances, altruistic behavior is an effect of various interactions between behavioral categories and cultural norms. Alexander (1987) regards altruistic behavior as an effect of systems of indirect reciprocity, where, in a society, it is costly for individuals to seem less altruistic than others. Here, the promotion of altruism can be seen as having an egoistic basis; in that it is in one’s own interest to surround oneself by altruists. This egoistic basis for promoting altruism is also important for understanding the discrepancy between the amount of preaching of altruism and the amount of actual altruistic behavior (Campbell, 1975). Both the promotion and the expression of altruism can have its root in group-egoistic tendencies of being conformist. In short, from an individual point of view, it is easy to understand why it is interesting to raise the general level of altruism in the population. However, it might be more difficult to understand why people let themselves be manipulated toward altruistic behavior. Is a more general inclination to conformism (Boyd & Richerson, 1991; Richerson et al., 2003) a sufficient explanation? One mechanism that may thwart regular group egoism into altruistic behavior is a slow process of increased imbalance between various interests, foremost those of the leader versus other members. When permanent
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
45
structures arise within a group, the link between members’ interests and the group’s policy becomes weaker since leaders and functionaries in the group acquire a self-interest that does not coincide with the ordinary member. The connection between interests and goals is no longer so obvious. A risk always exists that the group will degenerate, making leaders’ interest the main theme and turning members into just tools and no longer the beneficiaries. An altruistic goal may be used as an ideological cover for such a transition. War as a phenomenon can be understood as a male coalitional reproductive strategy, that is, a group-egoistic endeavor with deep roots in phylogeny (van der Dennen, 1995, 2007). Also, men who fight against an invading army of genocidal intent are acting rationally in terms of kin selection, when they risk their lives to save their families. While kin selection has seldom been a cause of war during recent millennia, quite a few wars – both offensive and defensive – can be viewed as acts of group egoism. Wars that include plundering may often be rational according to group egoism: the warriors take clear risks, but these are justified by the opportunity for booty. Group-egoistic offensive war finds an old example in the raiding voyages of the Vikings. In numerous other wars, the soldier has had little to win and his life to lose. It is not a question of defending his family, which has frequently stayed well out of danger, at least in one of the warring states. Nor do the soldier’s pay or chances of plunder provide a sufficient gain. Participation in such a war is clearly a form of altruistic behavior that originates in group egoism in prior societies. Similarly, regular cooperation may be thwarted, so that one actor mainly is the receiver and the other mainly the giver in a relationship. One part may be manipulated or in other ways deceived into believing that the relationship is reciprocal, but that it is good manners to stress the giving. One important point to be made is that in all cases, where unselfishness is involved, there is another part whose selfish needs are provided for; an altruistic interaction needs both a giver and a receiver. With an emphasis on the virtue of sacrifice, there is often a lack of scrutinizing whether the benefits are sufficient to motivate the sacrifices – nor who in fact are the beneficiaries.
MORAL CONFLICTS How do people divide time and resources among the behavioral categories that we have outlined (Fig. 1), tasks such as reading, dining with a friend, caring for children, community life, and blood donations? A ‘‘normal’’ person does many things, which can be classified as self-interest, thus
46
BIRGITTA S. TULLBERG AND JAN TULLBERG Action categories and moral rule spheres
ALTRUISM INTEGRITY
ES
M IS
H
O U
N
SE
U
P
LF
IS
EG
N
N TI O
PE RA
O CO
K
SE
RO
RE TE SE
LF
-IN
L
IN
G
ST
N
IO
T EC
S
RECIPROCITY
Fig. 1. Five Categories of Action and Three Moral Rule Spheres. Rules in All Three Spheres Will Affect All Five Categories of Action (see text for further explanation).
increasing the size of that category. Further, he or she performs a smaller number of actions, which qualify as unselfish. The amount of time and energy devoted to actions in the three other categories is likely intermediary between that of self-interest and unselfishness. Moral norms will influence human actions, and moral philosophy can be viewed as a dynamic conflict between norms. Specifically, conflicts may arise between the spheres that we envisage in this chapter (Fig. 1), but also between alternative norms within a particular sphere. Ethics will influence action partly within a behavioral category, and partly considering a choice between categories. In each sphere, there are alternative proposals for norms, but they have a unified tendency. The morality of integrity aims chiefly to strengthen individual action by opposing what it regards as bad actions in the other categories; for example, when an individual is forced into conventional behavior by her family, by ‘‘everyone else,’’ or by the state. Greater independence would lead to better consequences according to norms within the integrity sphere. Reciprocal morality relates to problems that could be solved to improve the advantage of cooperation. Finally, norms within the altruistic sphere stipulate that problems should be solved through individuals acting selflessly. Thus, each moral sphere should be taken to affect a broad spectrum of human actions not only in one specific behavioral category.
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
47
On numerous issues, conflict arises between a morality of integrity and a reciprocal morality. How free should individuals be in relation to their agreements: should one be able, for instance, to easily dissolve a contract of employment, marriage, or partnership? Where is the line distinguishing liberty from responsibility? Yet the two kinds of morality are connected by a common basis, self-interest in the broad sense. Generally, cooperation – as well as written or unwritten rules – is no sacrifice for the individual, but serves his or her own interests. The conflicts between these two spheres are more a question of border skirmishing than of total opposition. The classical conflict in moral philosophy is that between the integrity and the altruistic spheres. Because humans are regarded as basically selfish, altruism has to be taught and preached. Campbell illustrates the conflict between society and individual interests that has been resolved by societal preaching 100% altruism, but in reality manages to reach it to a much lower extent (Campbell, 1975). By necessity, such a moral system is hypocritical because people live up to the norms to such a low degree (Mackie, 1977). Is such a moral system necessary for a humane society? There is a third conflict, unnoticed by many, namely that between the reciprocal and the altruistic spheres. Whereas an altruistic morality is based on a one-sided giving (and one-sided receiving), a reciprocal morality is based on mutual gain, and this makes quite a lot of difference when it comes to ethical rules. Some citations from the Sermon on the Mount may serve as an illustration of an altruistic morality that directly confronts a reciprocal morality: ‘‘Love your enemies, do good to those who hate you, bless those who curse you, pray for those who abuse you. To him who strikes you on the cheek, offer the other also; and from him who takes away your cloak, do not withhold your coat as well. Give to every one who begs from you; and of him who takes away your goods, do not ask them again.’’ ‘‘If you love those who love you, what credit is that to you? For even sinners love those that love them. And if you do good to those who do good to you, what credit is that to you? For even sinners do the same.’’
Many social functions can proceed upon either a reciprocal or an altruistic foundation, but we believe that a fundamental hypothesis should be tested: Would we have a better society if an altruistic morality were replaced by a reciprocal one? Would it not be beneficial with fewer willing givers of blood on the battlefield, even if so also in the hospital? The implications are many and complex, but few if any issues have such importance for moral philosophy.
48
BIRGITTA S. TULLBERG AND JAN TULLBERG
As mentioned, there is no strict correspondence between the norm spheres and the categories of action. It is therefore important to clarify and exemplify how moral spheres and action categories combine. For example, a common combination is that between altruistic norms on one hand, and leadership egoism together with group egoism from the ordinary members on the other. The leadership might step by step change the policy so that there is no longer a group-egoistic rationale for the members, but their actions in effect become altruistic. The members may, however, believe that they are still the beneficiaries of the group project and that the altruistic norms only have a decorative function. Even the leader might be unaware of his egoistic advantages and firmly believe in himself sacrificing for a great altruistic purpose. Self-deception can be most helpful when trying to deceive others (Trivers, 1985). The relationship between leader and regular group members may in other instances be influenced by reciprocal norms under which the mutual benefits as well as responsibilities and costs are under steady scrutiny. When influenced by such norms, it should be more difficult for leaders to cheat and easier to keep them loyal to group-egoistic goals. A difference in interests is a source of conflict between different groups, but norms also influence such conflicts. A strong sense of integrity of one group confronts another group’s insistence of general obedience to a norm they honor. Conflicts can also occur between groups that have similar values. Two companies or two countries might both honor reciprocal values, but may experience the other part violating these norms. The potential cost one’s own group is always a factor to consider when judging the rationale of escalating the conflict. The problem with altruism is that this rationale is attributed less importance since self-sacrifice is held in such high regard. The confrontation between two groups, both considering themselves standing for high altruistic ethics, is likely to be especially antagonistic. In contrast, reciprocal ethics has a higher potential for solving conflicts between ethnic groups (Tullberg & Tullberg, 1997).
CONCLUSIONS Moral philosophy can be viewed as a dynamic conflict between norms and we envisage three moral spheres: (1) morality of integrity that argues for maximization of the individual’s freedoms and rights (self-interest), (2) reciprocal morality that promotes effectiveness in cooperation, and (3) altruistic morality that promotes sacrifices for others. What unites the
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
49
first two spheres is the individual’s wider self-interest since effective cooperation is of benefit to both parties. Thus, an overemphasis on morality of integrity at the expense of reciprocal morality can be disadvantageous to the individual in a long run or broader perspective since there should be an interest for the individual in achieving a balance between these spheres. Yet the altruistic sphere, advocated by traditional moral philosophy and religion, is diametrically opposed to the other spheres because of its focus on selflessness. Traditionally, altruism’s main impact is seen as that of counteracting egoism. However, there is opposition between the altruistic sphere, with its rules promoting a one-way giving, often departing from the needs of one party, and the reciprocal sphere, with its emphasis on mutual benefit. Reciprocal morality not only argues against parasitism and free riding for the self but also for others. Help is conditioned upon a reasonable prognosis of reciprocation by the receiving agent. Altruistic morality, on the other hand, is unconditioned, and there is no check against one party being used. This may be clouded by statements that generosity pays even in situations when this is not so. More attention ought to be paid to altruism’s potential subversive effect on reciprocal morality. Hypocrisy is generally and forcefully condemned, but what rightly can be called double standards is in effect presupposed by the agitation for an altruistic ethic of unrealistic ambitions. We maintain that morality should be required to be honest, containing rules that people can apply and have a serious ambition to apply (Mackie, 1977). This should be possible both for a morality of integrity and reciprocal morality.
ACKNOWLEDGMENTS We thank Olle Brick, Helena Cronin, Ingemar B Lindahl, Patrik Lindenfors, and Hanno Pichl for valuable discussions of previous versions of the manuscript. This work was supported by the Swedish Natural Science Research Council (to B.S.T.) and Svenska Handelsbanken (to J.T.).
REFERENCES Alexander, R. D. (1987). The biology of moral systems. New York, NY: Aldine de Gruyter. Alexander, R. D. (1989). Evolution of the human psyche. In: P. Mellars & C. Stringer (Eds), The human revolution: Behavioural and biological perspectives on the origins of modern humans (pp. 455–513). Edinburgh, UK: Edinburgh University Press.
50
BIRGITTA S. TULLBERG AND JAN TULLBERG
Axelrod, R. (1984). The evolution of co-operation. New York, NY: Basic Books. Axelrod, R., & Hamilton, W. D. (1981). The evolution of co-operation. Science, 211, 1390–1396. Barkow, J. H., Cosmides, L., & Tooby, J. (Eds). (1992). The adapted mind: Evolutionary psychology and the generation of culture. Oxford: Oxford University Press. Bastiat, F. (1848). L’Etat. Journal des De´bats, September 25, 1848. Bernhard, H., Fischbacher, U., & Fehr, E. (2006). Parochial altruism in humans. Nature, 442, 912–915. Boyd, R., & Richerson, P. J. (1991). Culture and co-operation. In: R. A. Hinde & J. Groebel (Eds), Co-operation and prosocial behavior (pp. 27–48). Cambridge, MA: Cambridge University Press. Brown, R. (1986). Social psychology: The second edition. New York, NY: The Free Press. Campbell, D. T. (1975). On the conflicts between biological and social evolution and between psychology and moral tradition. American Psychologist, 30(12), 1103–1126. Cialdini, R. B. (1988). Influence: Science and practice. Glenview, IL: Scott Foresman and Company. Cosmides, L. (1989). The logic of social exchange: Has natural selection shaped how humans reason? Studies with the Wason selection task. Cognition, 31, 187–276. de Waal, F. (1996). Good natured. The origins of right and wrong in humans and other animals. Harvard: Harvard University Press. Fehr, E., & G€achter, S. (2002). Altruistic punishment in humans. Nature, 415, 137–140. Fehr, E., & Henrich, J. (2003). Is strong reciprocity a maladaptation? On the evolutionary foundations of human altruism. In: P. Hammerstein (Ed.), Genetic and cultural evolution of cooperation (pp. 55–82). Cambridge, MA: MIT Press. Fessler, D. M. T., & Haley, K. J. (2003). The strategy of affect: Emotions in human cooperation. In: P. Hammerstein (Ed.), Genetic and cultural evolution of cooperation (pp. 7–36). Cambridge, MA: MIT Press. Frank, R. H. (1988). Passions within reason. New York, NY: W.W. Northon & Co. Gauthier, D. (1986). Morals by agreement. Oxford: Clarendon Press. Gintis, H., Bowles, S., Boyd, R., & Fehr, E. (2003). Explaining altruistic behavior in humans. Evolution and Human Behavior, 24, 153–172. Hamilton, W. D. (1964). The genetical evolution of social behavior. Journal of Theoretical Biology, 7, 1–52. Hamilton, W. D. (1971). Geometry for the selfish herd. Journal of Theoretical Biology, 31, 295–311. Hardin, G. (1993). Living within limits. Oxford: Oxford University Press. Hume, D. (1739). A treatise of human nature. London: Oxford University Press. Humphrey, N. (1997). Varieties of altruism – and the common ground between them. Social Research, 64(2), 199–209. Huxley, T. H. (1894). Evolution and ethics. London: MacMillan and Co. Krams, I., Krama, T., & Igaune, K. (2006). Mobbing behavior: Reciprocity based co-operation in the breeding Pied Flycatcher Ficedula hypoleuca. Ibis, 148, 50–54. Lopreato, J. (1981). Toward a theory of genuine altruism in Homo sapiens. Ethology and Sociobiology, 2, 113–126. Mackie, J. L. (1977). Ethics, inventing right and wrong. Harmondsworth, London: Penguin Books. Masters, R. D. (1983). The biological nature of the state. World Politics, 35, 161–193.
Integrated Evolutionary Perspective on Normative Ethics and Human Behavior
51
Masters, R. D. (1989). The nature of politics. London: Yale University Press. Maynard Smith, J., & Price, G. R. (1973). The logic of animal conflict. Nature, 246, 15–18. Pinker, S. (2002). The blank slate. London: Penguin Books. Richerson, P. J., Boyd, R. T., & Henrich, J. (2003). Cultural evolution of human cooperation. In: P. Hammerstein (Ed.), Genetic and cultural evolution of cooperation (pp. 357–388). Cambridge, MA: MIT Press. Ridley, M. (1996). The origin of virtue. New York: Viking. Ruse, M. (1986). Taking Darwin seriously: A naturalistic approach to philosophy. Oxford: Basil Blackwell. Ruse, M. (1989). The darwinian paradigm: Essays on its history, philosophy, and religious implications. London: Routledge. Sober, E. (1993). Philosophy of biology. Oxford: Oxford University Press. Sober, E., & Wilson, D. S. (1998). Unto others – The evolution and psychology of unselfish behavior. Cambridge, MA: Harvard University Press. Swap, W. C. (1991). Perceiving the causes of altruism. In: R. A. Hinde & J. Groebel (Eds), Co-operation and prosocial behavior (pp. 147–158). Cambridge, MA: Cambridge University Press. Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology, 46, 35–57. Trivers, R. L. (1985). Social evolution. Menlo Park, CA: Benjamin/Cummings. Tullberg, J. (2003). Rationality and social behavior. Journal of Theoretical Biology, 224, 469–478. Tullberg, J., & Tullberg, B. S. (1997). Separation or unity? A model for solving ethnic conflicts. Politics and the Life Sciences, 16, 237–248. van den Berghe, P. (1981). The ethnic phenomenon. New York: Praeger. van der Dennen, J. M. G. (1995). The origin of war – the evolution of a male-coalitional reproductive strategy. Groningen, The Netherlands: Origin Press. van der Dennen, J. M. G. (2007). Three works on war. Politics and the Life Sciences, 26, 71–92(Book Review). Wilkinson, G. S. (1984). Reciprocal food sharing in the vampire bat. Nature, 308, 181–184. Williams, G. C. (1966). Adaptation and natural selection. Princeton, NJ: Princeton University Press. Williams, G. C. (1989). A sociobiological expansion of evolution and ethics. In: J. Paradis & G. C. Williams (Eds), Evolution and ethics: T. H. Huxley’s evolution and ethics with new essays on its victorian and sociobiological context. Princeton, NJ: Princeton University Press. Wilson, E. O. (1975). Sociobiology: The new synthesis. Cambridge, MA: Harvard University Press. Wilson, E. O. (1979). On human nature. New York: Bantam Books. Wilson, E. O. (1998). Consilience: The unity of knowledge. New York, NY: Knopf. Wright, R. (1994). The moral animal. New York: Pantheon.
THE BIOPOLITICS OF PRIMATES$ Johan M. G. van der Dennen INTRODUCTION In this chapter, I use the term ‘‘biopolitics’’ to mean evolutionarily informed political science. Politics has been characterized as ‘‘Who gets what, when, and how’’ (Lasswell, 1936), but rather than about material possessions, politics is understood to be about power, more specifically about collective power, especially differential group power competition, hierarchy and stratification in power distribution, and the universal struggle to enhance power, and to maintain or challenge/destroy this status quo. Politics ‘‘should be found in any system of nature in which conflicts of interest exist among cooperating organic units’’ (Johnson, 1995, p. 279). My main focus will be competitive intergroup relations in monkeys and apes, or as I (van der Dennen, 1995) called it ‘‘intergroup agonistic behavior’’ (IAB). I also briefly treat interindividual and intercoalitionary agonistic behavior when relevant.
PRIMATES IN THEIR NATURAL HABITAT: THE FIRST GENERATION OF PRIMATOLOGISTS Yerkes and Yerkes (1929) had been studying the behavior of individual monkeys for many years in America. Under their auspices, Carpenter went $
Dedicated to the memory of Adriaan Kortlandt, animal psychologist, who died in 2009.
Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 53–96 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009005
53
54
JOHAN M. G. VAN DER DENNEN
to the island of Barro Colorado in Central America. Here in 1931–1933, he studied the societies of howler monkeys (Alouatta palliata). Carpenter also made field studies of red spider monkeys (Ateles geoffroyi) in the Coto region on the borders of Panama and Costa Rica; and in 1937 he was watching gibbons (Hylobates lar) on the hillsides of Thailand. The Cayo Santiago colony has been studied by Carpenter himself (1939–1940), Altmann (1956–1958), and Koford (1958 onward). Soon after the end of the Second World War, enterprising Japanese scientists (notably the Primate Research Group of Kyoto University) began to study the monkeys on the islet of Koshima and on Takasakiyama. In 1956 the Japan Monkey Center was set up at Inuyama. Together with other early observations by (in alphabetical order): Altmann, Bygott, Carpenter, Cheney, DeVore, Fossey, Galdikas, Gartlan, Goodall, Haddow, Hall, Hamburg, Hrdy, Itani, Izawa, Jay, Kawanaka, Koford, Kortlandt, Kummer, Mason, Moynihan, Nishida, Nissen, Reynolds, Saayman, Schaller, Seyfarth, Silk, Southwick, Struhsaker, Sugiyama, Suzuki, Teleki, Ullrich, Washburn, Wrangham, among many others, these studies revealed the intricate societal structures, the affiliative systems, and the rather peaceful modus vivendi (with some spectacular exceptions) of monkey and ape populations. Violence, indeed, appeared to be a density-dependent phenomenon in captive populations, enhanced by an unbalanced sex ratio. Also early observations in the wild suggested that chimpanzees (Pan troglodytes) lived in an amiable web of friendships and occasional get-togethers (Nissen, 1931; Reynolds & Reynolds, 1965; Goodall, 1968, 1971; Albrecht & Dunnett, 1971). The apes were now regarded as gentle, peace-loving creatures, especially after the publications on bonobos (Pan paniscus) by Parish (1994, 1996), de Waal (1984), and Parish and de Waal (1999). In retrospect, the 20 most important discoveries in primatology from a biopolitical and ‘‘biocriminological’’ perspective (not in chronological order) are: 1. Discovery of infanticide in many primate species (Bygott, 1972; Hrdy, 1977a, 1977b, 1979; Hausfater & Hrdy, 1984; van Schaik & Janson, 2000; Goodall, 1977, 1986). 2. Discovery of carnivority: meat eating and (cooperative communal) hunting in chimpanzees by Goodall and her team in 1960; Goodall (1977, 1986); McGrew Tutin, Baldwin, Sharman, and Whiten (1978) and
The Biopolitics of Primates
3.
4.
5.
6. 7. 8. 9.
55
McGrew, Tutin, and Baldwin (1979); meat eating has also been observed in bonobos (Ihobe, 1992). Discovery of cannibalism in chimpanzees (Suzuki, 1971; Bygott, 1972); Goodall: Passion and Pom. A female chimpanzee Passion and her daughter Pom were seen to kill and eat three Kasakela infants and almost certainly killed seven more newborns over a period of four years (Miller, 1995, p. 107). Discovery of violence against disabled (polio) and ‘‘alien’’ chimpanzees (ethnocentrism and xenophobia: Goodall); Goodall (1971, et seq.) reported the case of a polio-stricken chimpanzee, partly paralyzed, who was being ostracized, mobbed, and harassed by the other chimpanzees [Goethe (1939) designated this kind of behavior as ‘‘Anstossnehmen’’]. Bygott (1979) suggested that there would be strong selection for chimpanzee males to be rapidly aroused to attack strangers, particularly males, on sight. Wilson and Wilson (1968) noted that in a captive chimpanzee colony, the most severe attacks were directed at new individuals who had just been introduced into the enclosure and aggression against strangers and introducees has been noted in many other primate studies (e.g., Hall, 1962; Southwick, 1967; see review in van der Dennen, 1995) in the wild, as well as in the semi-captive colony at the Arnhem Zoo (de Waal, 1982). Especially in relation to human ethnocentrism and xenophobia, it is particularly interesting that in the nonhuman primates generally – as in humans – intragroup cohesion and intergroup hostility may be correlated (Cheney, 1992). Discovery that chimpanzees attack an animated stuffed leopard with sticks and clubs (Kortlandt, 1965, 1975, 1999). ‘‘Common chimpanzees can throw projectiles with an overarm movement and use big sticks as clubbing weapons against a stuffed leopard model. Surprisingly, nothing of this kind seems ever to have been observed in the gracile chimpanzee’’ (i.e., the bonobo) (Kortlandt, 1999, p. 35). Discovery of degrees of female ‘‘promiscuity’’ and consortship (‘‘honeymooning’’) in chimpanzees. Discovery of rape (coercive or forcible copulation) in orangutans (Pongo pygmaeus) (Pitcairn, 1974; Galdikas, 1979; MacKinnon, 1979; among others). Discovery of ‘‘battle-type’’ warfare in baboons, macaques, and other primate species (Cheney, van Hooff, Pusey, Hrdy, Smuts, among others). Discovery of ‘‘raiding-type’’ ‘‘primitive warfare’’ in chimpanzees (Goodall, Wrangham, Nishida, among others). Lethal male raiding indicate ‘‘an appetite for hunting and killing rivals that is akin to predation’’
56
10.
11.
12.
13.
JOHAN M. G. VAN DER DENNEN
(Wrangham, 1999, p. 5). Van der Dennen (1995, p. 171) already argued that ‘‘battle-type’’ warfare and ‘‘raiding-type’’ warfare corresponded with two discernible motivational systems, the one tilting toward fight/ flight motivation, the other toward predation. Discovery of females as ‘‘warriors’’ in a number of female-philopatric species; female participation in IAB [e.g., Macaca mulatta: Hausfater, 1972; Hanuman or gray langurs [Presbytis (Semnopithecus) entellus]: Ripley, 1967]. Social behavior of muriquis or woolly spider monkey (Brachyteles arachnoides). Despite their evolutionary and ecological distance from the apes, muriquis nonetheless exhibit striking similarities in various aspects of their social behavior. Most notable are the highly affiliative, egalitarian relationships that philopatric males maintain with one another and with females in their groups (Strier, 2001, p. 80); Wrangham (2006) called the muriqui ‘‘Perhaps the most dramatic primate example of a non-violent species.’’ Strier, Ziegler, & Wittwer (1999) proposed that both the lack of significant changes in testosterone levels with the onset of the copulatory season and the lack of prebreeding increases in cortisol may be related to the low levels of overt aggression displayed by male muriquis over access to mates (see also Strier, 2001, et seq.). Although aggressive interactions were rare among group members, the males showed strong solidarity when they took a stance in encounters against males from neighboring muriqui groups (Strier, 2001, p. 81). A striking feature of muriqui mating patterns is that males make no overt moves to interfere with one another’s sexual activities and do not try to harass females into copulating with them. Instead, males wait patiently until a receptive female favors them with a chance to mate. Fighting among males is futile in the face of unmonopolizable mates, so muriqui males rely for fertilizations on female favoritism and perhaps on more subtle forms of competition (Strier, 2001, p. 84). Discovery of raiding behavior in male spider monkeys (Ateles geoffroyi yucatanensis) (Aureli, Schaffner, Verpooten, Slater, & Ramos-Fernandez 2006), revisiting the debate on the relationship between cooperative hunting and warfare. Like chimpanzees, spider monkeys are characterized by male philopatry, a high degree of fission–fusion dynamics, strong associations between males, and male cooperative territorial defense. Discovery of sex as reconciliation in bonobos (Parish, 1994, 1996; de Waal, 1984; de Waal & Lanting, 1998), and the debate over which Pan-species can serve as model for human phylogeny: ‘‘Current debates
The Biopolitics of Primates
14.
15.
16.
17.
18.
19.
20.
57
are therefore not about which species is closer to us, the bonobo or chimpanzee, but rather which one more resembles the last common ancestor’’ (de Waal, 2001, pp. 44–45). Discovery of empathy, sharing, reconciliation, reassurance behavior, consolation, compassion, ‘‘morality,’’ etc. (e.g., de Waal & van Roosmalen, 1979; Preston & de Waal, 2003; Arnold & Aureli, 2006; de Waal, 2010). Discovery of many sexual dimorphisms in behavior paralleling those in humans, including dominance, rough-and-tumble play (e.g., Dolhinow & Bishop, 1972; Hamburg & Goodall, 1974), and even juvenile play in vervet monkeys [Cercopithecus (Chlorocebus) aethiops] and rhesus macaques (experiments by Alexander & Hines, 2002). Discovery of coalitionary killing in the western red colobus monkey (Colobus badius). At least two and possibly four coalitionary kills were recorded by Starin (1994). In these cases, coalitions of females attacked and killed males attempting to enter their groups (Wrangham, 1999, p. 5). Before this discovery only male chimpanzees in the wild (Goodall, 1979, 1986; Goodall et al., 1979) and in semi-captivity (de Waal, 1982) had been known to use coalitionary lethal power politics. Discovery of primate hunting and meat eating in bonobos (Surbeck & Hohmann, 2008; Surbeck, Fowler, Deimel, & Hohmann, 2009). Before this time, bonobos were supposed not to hunt communally, and, especially, not to hunt monkeys. Discovery of ‘‘incipient culture’’: tool making and tool use traditions in (mainly female) chimpanzees (by Goodall and her team in 1960) and orangutans (e.g., Goodall, 1970, 1986; McGrew,1992; McGrew, Marchant, & Nishida, 1996; Nishida, 1987). Discovery of spear-like implements for hunting by female savanna chimpanzees (P. troglodytes verus) in Fongoli (Senegal), where high levels of in-group and out-group directed aggression has not been observed (Pruetz & Bertolani, 2007); once again questioning the link between hunting and conspecific killing. Social deprivation experiments in rhesus monkeys by Harlow, Suomi, Sackett, and colleagues (Harlow’s Wisconsin deprivation research): Early isolation in young monkeys leads to hyperaggressiveness, selfmutilation, hypersexuality and bizarre sexual behavior, and pathological social behavior (e.g., Deets & Harlow, 1971). Motherless mothers: Monkeys raised without their mothers and other conspecifics show inappropriate aggression, making poor social companions, and as a
58
JOHAN M. G. VAN DER DENNEN
result, when placed into social settings, they are not preferred as social partners. They are generally shunned by other monkeys, and end up low in social dominance rank (Higley, 2003, p. 26). Biopolitics and primatology have been bedfellows from the very beginning of their disciplines (e.g., Masters, 1975, 1983, 1989; Willhoite, 1976; Peterson & Somit, 1980; Schubert, 1983, 1986 1991; Somit, 1990; Schubert & Somit, 1982; Schubert & Masters, 1991; G. Johnson, 1995; Somit & Peterson, 1995, 1998).
Interindividual Agonistic Behavior Threat, attack, and submissive behavior patterns (together termed ‘‘agonistic behavior’’) are most likely to occur among chimpanzees under the following conditions (Goodall, 1968, 1971, 1986; Hamburg, 1972, 1978; Bygott, 1979; Tutin, 1975; Pusey, 1979; Trudeau, Bergmann-Riss, & Hamburg, 1981): (1) in daily interactions involving dominance (status); (2) following the recent occurrence of other attacks; (3) in long-term changes in status or dominance, particularly among males; (4) when apparent unexpressed aggression toward higher ranking members is redirected at a lower status individual; (5) in the protection of infants by adults of both sexes, but especially by females; (6) in defending against potential predators; (7) in killing and eating animals of other species; (8) in terminating severe disputes among subordinate animals; (9) in association with a presumably painful injury; (10) in exploration of strange or threatening areas; (11) in meeting relatively unfamiliar chimpanzees; (12) in circumstances where highly valued resources are in short supply; (13) by resident females toward newly transferring females; and (14) when relatively unfamiliar animals are crowded in the presence of highly valued resources. Additionally, Bygott (1979) reports that the most frequent context of attacks and displays occurred at meetings between two individuals who had been separated for at least half an hour. Some further generalizations on agonistic relationships offered by Bygott (1974) are (1) a male’s ability to dominate other males in agonistic interactions changes with age, peaking at physical maturity; (2) attacks between adult males of the same community are very rare; (3) adult males are dominant in all their fights with all other age/sex classes; (4) two or more adult males may form a coalition (usually between brothers) that strengthens the status of both members over that which could have been obtained
The Biopolitics of Primates
59
individually; (5) agonistic interactions are rarer than expected between relatives of different age/sex classes; and (6) agonistic relationships between females are rare. Now, such patterns and contexts of agonistic behavior may be taken to be fairly representative for the primates in general, though every species shows its own species-specific profile of agonistic patterns and contexts. A characteristic form of agonistic behavior lacking in chimpanzees, is, for example, the ‘‘stink fight’’ of ring-tailed lemurs (Lemur catta) (Jolly, 1966), the aggressive ‘‘herding’’ of females as described in Papio [c.] hamadryas (Hall & DeVore, 1965; Kummer, 1968), or violent rape as described in the orangutan (e.g., Pitcairn, 1974). Furthermore, primate species differ in their degree of showing breeding seasonality of aggression, different types of dominance hierarchies, sexual harassment, territoriality, coalitions, social organization, and so on. But, by and large, the agonistic repertoire as described above gives a fairly accurate picture of primate agonistic behavior in general. Bernstein and Ehardt (1985) reviewed data from a wide variety of species and report that aggression, on average, accounts for 2–5% of a monkey’s daily activity budget (Higley, 2003, p. 19). The most frequent context in which aggression is seen is in defense of status (i.e., dominance) (Higley, 2003, p. 20; Walters & Seyfarth, 1987, pp. 307–308). In both captive and free-ranging animals, old scars and wounds, at times even severe wounds, are frequently observed, especially among adult males (Rawlins & Kessler, 1986; Southwick, Beg, & Siddiqi, 1965; Steenbeek Piek, van Buul, & van Hooff, 2000). Females also receive wounds, especially during the breeding season (Higley, 2003, pp. 20–21). The social setting in which aggression is most likely to occur is probably when two troops meet, or, in captivity, during the introduction of unfamiliar monkeys into an existing group, or the formation of a new group (Altmann, 1962; Bernstein, Gordon, & Rose, 1974; Carpenter, 1974; Cheney, 1981; Poirier, 1974; Southwick et al., 1965; Southwick, Siddiqi, Farooqui, & Pal, 1974; Higley, 2003, p. 22; see review in van der Dennen, 1995). While severe aggression is more frequent between strangers, among some particularly aggressive individuals, and within groups of some species, it occurs among virtually all primate species studied. While it is often believed that primate males are more aggressive than females, among many species of Old World primates, rates of aggression are surprisingly similar between the sexes. Nevertheless, some sex differences have been demonstrated (Higley, 2003, p. 24). Socially living primates must learn not only to recognize social cues of aggression but also to restrain and control their own impulses whenever
60
JOHAN M. G. VAN DER DENNEN
necessary. In fact, achieving high dominance status within a troop may depend on inhibiting aggression as much as on expressing it (Higley, 2003, p. 26). The frequency of aggression between two individuals does not necessarily reflect the overall aggressiveness of their social relationship. Since rates of aggression are to some extent dependent on the frequency with which individuals come into contact, animals that associate regularly may appear to be more aggressive toward each other than ones that spend less time together. For example, same-sexed peers and kin often show high rates of aggression. However, these individuals usually interact less aggressively than others when the frequency of association is taken into account [e.g., toque macaques (Macaca sinica): Dittus, 1977; Japanese macaques (Macaca fuscata): Kurland, 1977]. Similarly, in species with one-male groups, male– male aggression may occur relatively infrequently simply because there is little opportunity for such interaction. Where males do come into contact, however, their interactions are often extremely hostile [e.g., blue monkeys (Cercopithecus mitis): Cords, 1987; redtail monkeys (Cercopithecus ascanius): Cords, 1984, 1987; black-and-white colobus (Colobus guereza): Oates, 1977; gray langurs: Hrdy, 1977a, 1977b). In multimale groups of baboons, aggression between adult males does not always occur at high rates, but since aggression often accounts for an extremely high proportion of their interactions, their relationships appear to be relatively hostile (Walters & Seyfarth, 1987, p. 308). Infanticide It is now well documented that Hanuman langurs, after driving a leader male from his females, may systematically kill all infants in the troop (Sugiyama, 1967; Mohnot, 1971; Hrdy, 1974, 1977a, 1977b, 1979). In this way the new leader ensures that most subsequent infants will carry his genes. Of interest are Fossey’s (1979) observations relating to infanticide in the mountain gorillas: three infants were killed as a result of aggressive intergroup interactions. Moreover, on one occasion the lone silverback responsible left with the mother of the dead infant a week later. Killing of infants has also been reported in chimpanzees (Suzuki, 1971; Bygott, 1972; Goodall et al., 1979), patas monkeys (Erythrocebus patas) (DeVore, personal communication, in Fossey, 1979), purple-faced langurs (Presbytis senex senex) (Rudran, 1973), redtail monkeys (Struhsaker, 1977), € & Angst, 1974), crab-eating macaques hamadryas baboons (Kummer, Gotz, (also known as Java monkeys) (Macaca fascicularis) (Washburn & Hamburg, 1968), and Verreaux’s sifaka (Propithecus verreauxi verreauxi) (Littlefield, 2010).
The Biopolitics of Primates
61
Infanticide and cannibalism often go together, for example, in chimpanzees (Bygott, 1972; Goodall, 1977, et seq.; Kortlandt, 1980; Nishida, 1980; Kawanaka, 1981; Itani, 1982; Schubert, 1983). Confusing paternity to mitigate male aggression is thought to be a common female anti-infanticidal tactic (Strier, 2001, p. 88). Sexual Coercion and Harassment On sexual coercion in primates see Muller and Wrangham (2009). The most plausible alternative to the fertilization tactic theory of rape is the sexual coercion hypothesis proposed by Smuts and Smuts (1993; Clutton-Brock & Parker, 1995). In some species, according to this line of thought, rape may be an evolved male mechanism whose primary aim is not fertilization in the present, but control – for the ultimate purpose of fertilization in the future. Sexual coercion looks also to be the underlying reason of male battering of female chimpanzees, because the mating patterns in Gombe demonstrate how particularly effective male domination is. As Goodall (1986, p. 481) described it: ‘‘Almost always, unless he is crippled or very old, an adult male can coerce an unwilling female into copulating with him.’’ So the rule seems to be: Coercion works. Furthermore, Goodall stated: ‘‘Males may attack females seemingly in order to drum into their victims, again and again, that theirs is a male-dominated society’’ (see also Silk, 2002a). Many primate females have active tendencies toward polyandrous mating and show opportunistic sexual proceptivity when encountering new males. Harassment by males of sexually attractive females is most salient among the Old World primates, especially in species in which males are much bigger than females or females range alone at least some of the time. Possessive mate guarding is also most pronounced in this taxon, as is the incidence of males injuring females. In other primates, females seem to have much more control over their mating behavior, including the identity of their mating partners. It is therefore likely that unusual sexuality of catarrhine females evolved in part to reduce the risk of infanticide (van Schaik, 2002, p. 938). Reconciliation and Consolation Reconciliation (e.g., Silk, 1997, 1998, 2002b; de Waal, 2006) should not be exaggerated. For example, Daniel, Santos, and Cruz (2009) recorded 190 agonistic interactions and subsequent postconflict behavior in a captive group of brown capuchin monkeys (Cebus apella). Only 26.8% of these conflicts were reconciled. Reconciliation was more likely to occur between
62
JOHAN M. G. VAN DER DENNEN
opponents that supported each other more frequently and that spent more time together (supporting the ‘‘valuable relationship’’ hypothesis). One can deduce from de Waal’s work that reconciliation and ‘‘otherregarding behavioral dispositions’’ are only an in-group phenomenon. Adang (1999) has emphatically pointed out that there is no – never – reconciliation after episodes of intergroup violence in chimpanzees.
Opportunistic Coalitions Over the past decades a continuing study of a (semi-)captive group of chimpanzees at the Arnhem Zoo in the Netherlands has focused on the importance of (opportunistically) coalitional behavior by males in the establishment and maintenance of, and change in, the primary (i.e., the male) dominance structure of the group (van Hooff, 1973a, 1973b; de Waal, 1978a, Adang, 1999). An independent female dominance structure is much more stable (see also McGuire, 1982; Hrdy, 1981). Dutch ethologists (No€e, de Waal, & van Hooff 1980; de Waal & van Hooff, 1981; de Waal, 1982, et seq.) and American political scientists alike have discussed the dominance behavior of the Arnhem chimpanzees in terms of cost–benefit analysis, and explicitly as political behavior homologous to that readily observable among contemporary humans (e.g., Schubert, 1983). It appears that rather than staying with relatives, adult males are opportunistic in their relationships, making and breaking alliances for individual advantage as the relative power of each male waxes and wanes (Pusey, 2001, p. 24). Adult males of many primate species frequently form alliances to dispute existing dominance relationships. This occurs, for example, when two males challenge a third who outranks them both [squirrel monkeys (Saimiri sciureus): Baldwin & Baldwin, 1981; baboons: Hall & DeVore, 1965; Packer, 1977; Rasmussen, 1981; Collins, 1981; Japanese macaques: Stephenson, 1975; chimpanzees: Goodall, 1968; de Waal, 1982; Nishida, 1983). Although such alliances often involve unrelated males, they may consistently involve the same individuals (e.g., Packer, 1977; Nishida, 1983; Walters & Seyfarth, 1987, p. 316). Plavcan, van Schaik, & Kappeler (1995) found that, among primate species in which the outcome of fights is typically determined by coalitionary fighting, selection for weaponry (canines) is reduced. In particular, male chimpanzees have relatively small canines for their body size, consistent with
63
The Biopolitics of Primates
other evidence that success in battle for chimpanzees depends on coalition size rather than individual weaponry (Wilson, Britton, & Franks, 2002; van Schaik, 2002, p. 937).
Polyadic Social Aggression Monkeys and apes often resolve fights based on the willingness of others to join on one side or another. In this way size and strength become much less important than the size and reliability of social alliances in resolving fights between individuals. It is no longer the largest and most powerful that will be the victor, but now perhaps the most socially adept (Bernstein, 1999, p. 74). Chapais (1991, pp. 200–202) distinguished six functional categories of (triadic, polyadic) aggressive alliances: protective alliances, revolutionary alliances, conservative alliances, resource-specific coalitions, defensive coalitions, and xenophobic alliances, which include all instances in which individuals belonging to the same group jointly threaten or attack one or more members of another group. Such coalitions may develop in three contexts: territorial defense, the repelling of potential immigrants, and dominance interactions between groups. Note that this category of xenophobic alliance is exactly the same as my category of IAB. Coalitions have the character of transactions that continue as long as both participants derive net benefits and/or have more influence or freedom of movement than in other possible combinations. In an analysis of coalition formation and agonistic third-party intervention in a semi-captive group of chimpanzees, de Waal (1982) distinguished a category of protective support, in favor of the weaker party in a conflict, and a category of opportunistic support, in favor of the stronger party. An increasing number of experimental studies provide evidence that dyadic social relationships in primate groups are affected by third group members and the relations among these (e.g., Kawai, 1965; Kummer et al., 1974). With respect to agonistic interactions complexity appeared to be the rule rather than the exception in two well-established captive groups of Java monkeys: in each group more than half of all aggressive actions occurred during interactions with three or more participants. A detailed description and categorization of the types of agonistic interaction among Java monkeys has been given by de Waal, van Hooff, and Netto (1976).
64
JOHAN M. G. VAN DER DENNEN
Two of these types of polyadic interactions have in common that one or several group members intervene aggressively in a conflict between others: (1) Agonistic interactions in which the aggressive party is attacked are called reactor-alliances (RA) and the intervening individual is referred to as a protective-aggressor. (2) If it is the aggressive party that receives support, the interaction type is called an actor-alliance (AA), in which we can distinguish the role of start-aggressor and one or several join-aggressors (de Waal, 1978b). From each group of monkeys about 1,000 agonistic interactions were recorded. The analysis of the data revealed that the tendency to join aggressors is considerably higher than the tendency to protect aggressees. In one group 135 AAs and 44 RAs occurred, in the other group 110 AAs and 53 RAs. The more altruistic form of support, which depends on a relationship of attachment and affiliation, and the more opportunistic kind, where intervention depends on the possible advantage gained by the supporter, have also been found in other primate species (e.g., Netto & van Hooff, 1986; see Harcourt & de Waal, 1992). Triadic and polyadic interactions appear to be common in primates generally (Walters & Seyfarth, 1987, p. 309). The paramount majority of all agonistic interactions is apparently an intragroup phenomenon, but some species also exhibit IAB that may be more or less collective and orchestrated. This will be the subject of the following sections.
Battles Observing hamadryas and gelada (Theropithecus gelada) baboons in Abyssinia, Sanderson (1955) witnessed ‘‘terrific battles, amounting almost to organized warfare – with surprise raids, the taking of prisoners, wide maneuvers, and other grossly human tactics.’’ Van Hooff (1990a, 1990b) vividly portrayed a massively escalated agonistic episode between two olive baboon (Papio [c.] anubis) groups of about 100 and 150 individuals respectively, at Gilgil, Kenya. He described it as an impressive, sometimes earpiercing, spectacular, and cacophonous event, lasting more than an hour. A great many animals from both groups formed two, more than 100-m-long frontlines, mutually threatening and making small sorties, meanwhile alternately ‘‘jerk-looking’’ at their own neighbors (for reassurance) and at the adversaries, and making a hell of
The Biopolitics of Primates
65
a noise (‘‘shriekbarking’’). Suddenly, presumably when one or more adversaries shrank and retreated, a number of animals simultaneously rushed forward in a massive assault. Almost immediately the entire frontline dashed forward and drove the other party before it for some 100 meters, accompanied by a crescendo of shriekbarking by the entire chasing group. Gradually the chase lost impetus, slowed down, the frontlines coming to a standstill and again the parties confronted each other. After a while a similar surging attack would be repeated in the same or the reverse direction, the groups chasing each other to and fro. The end of the confrontation was less spectacular: it ended not in a decisive final chase, but petered out because gradually less and less individuals participated in the forays. This was especially true for the intruding group, which finally withdrew to its own home area. Most notable, according to van Hooff’s account, was the small amount and short duration of physical contacts during the fight. There were no visible injuries, which was probably more due to the animals’ fear to be engulfed by the opponents, rather than to any magnanimous restraint or inhibition. Furthermore, there were no indications that the manifestly synchronized action could be ascribed to one or more coordinating leaders. It was also striking that the females were just as active in the skirmishes, if not more so, than the males (see also Wrangham & Peterson, 1996). DeVore and Hall (1965), describing intergroup relations in chacma baboons (Papio [c.] ursinus) of Nairobi Park and Amboseli, indicated that in many cases groups would drink at the same waterhole with no indication of disturbance. However, groups were not always equally ‘‘relaxed’’ when near each other. Fighting between groups trying to settle in the same sleeping trees has been seen at Amboseli (DeVore & Hall, 1965). Fighting between bands of hamadryas baboons – the most ‘‘sexist’’ society among the primates – is also conducted by the males. It consists almost entirely of spectacular bluffing, during which the opponents fence at each other with open jaws and slap swiftly back and forth with their hands. Film analysis showed that despite appearances, physical contact seldom occurs. Only when one male turns and flees, he is apt to receive a scratch on the anal region (Kummer, 1968). Kummer (1968) described a battle royal between bands of hamadryas over bait he provided. Recently (2009), two troops of hamadryas baboons have been filmed going to war, with hundreds of monkeys entering into a pitched battle (http://news.bbc.co.uk/earth/hi/earth_news/newsid_8400000/8400019.stm). The fight, filmed by the BBC Natural History Unit, appeared to be triggered by male baboons attempting to steal females from the harems of rivals. Usually, the two troops live relatively peacefully alongside one another on a 1-km-long cliff in the Awash National Park in Ethiopia. In baboons, as well
66
JOHAN M. G. VAN DER DENNEN
as other primate species, male individuals may participate in intergroup encounters merely to survey the mating opportunities in another group (Kitchen & Beehner, 2007, p. 1565). According to Kitchen and Beehner (2007, p. 1552), primate groups do not always, or even often, act as cohesive, cooperative units against extra-group animals, even when all members would benefit from excluding rivals. How individuals respond when facing extra-group rivals can vary within a species and even among members of the same group. Cases in which males cooperate to defend a group or territory are rare among mammals in general (Pusey, 2001, p. 19).
Raids Raids into neighboring territories may occur for different reasons, including the increase of foraging and mating opportunities directly or indirectly through the killing of neighboring rivals. Lethal raids have been mainly observed in humans and chimpanzees, with raiding males being reported to search purposefully for neighbors (for a more detailed treatment of this subject, see van der Dennen, 2011). Bygott (1979), Wrangham (1979), Goodall (1979), Goodall et al. (1979), Nishida (1979, 1980), and Itani (1982) were the first to report on the intercommunity relationships of the Gombe (Tanzania) population of chimpanzees (Pan troglodytes schweinfurthii), especially episodes of what Goodall literally called ‘‘primitive warfare.’’ Parties of up to ten adult males, sometimes accompanied by females and subadults, quite regularly patrol the boundaries, keeping close together, silent and alert, often stopping to listen intently, apparently actively searching for signs of neighbors. Sometimes they climb a tree to scout the ‘‘hostile’’ territory of the adjacent community, just like a human reconnaissance party might do (the original community had begun to divide into two separate communities about 1970). If no members of the neighboring community are detected, the patrol may stealthily intrude into the ‘‘enemy’’ territory. When a fairly large ‘‘enemy’’ party is encountered both parties may engage in vocal and gestural agonistic displays, or one of them may charge and chase the other away, or both give up and return to their core areas. At other times, a party, upon spotting ‘‘enemies,’’ may flee, thus avoiding encounter. When, however, small parties or single ‘‘enemy’’ chimpanzees, particularly anestrous females, are encountered by the ‘‘warriors,’’ these may be severely and viciously attacked and killed. Goodall (1979) described several such lethal episodes in some (gruesome) detail.
The Biopolitics of Primates
67
‘‘It seems,’’ she continued, ‘‘that we have been observing a phenomenon rarely recorded in field studies – the gradual extermination of one group of animals by another, stronger, group. Why these brutal attacks? The northern males were not defending their own territory, since all the attacks except one were deep within the southern community home range. On the other hand, the aggressor males, before the community split, had access to the area that the southern community took over. If they were merely trying to reclaim territory they had lost, then they have certainly succeeded.’’ Subsequently, Goodall (1986) reported observations of five lethal attacks, and some 13 more that left the victims – including adults and infants of both sexes – severely wounded and bleeding profusely. Why, she wondered, would the aggressors attempt to kill, maim, or injure their victims instead of merely chasing them away? By the end of the Four Year War, the Kahama community – seven males and three adult females and their young – had been annihilated. Researchers witnessed five of the attacks, in which Kasakela chimpanzees tore at their victims’ flesh with their teeth as if they were common prey (Miller, 1995, p. 106). From the five long-term sites where chimpanzees live with neighbors (Budongo, Gombe, Kibale, Mahale, and Taı¨ ) we now have more than 180 years of data on 11 communities. The 5 sites yield direct evidence of kills of neighbors in 33 cases, and a further 16 suspected victims. They also show repeated interest in killing, evidenced from patrolling of borders to attempts to isolate, immobilize, and intensely batter or cut a victim (Wrangham, 2006). Aureli et al. (2006) reported on the first cases ever witnessed of raiding parties of male spider monkeys (Ateles geoffroyi yucatanensis), a species expected to show such a behavioral tendency, given its similarity with humans and chimpanzees in critical socio-ecological characteristics, such as fission–fusion social dynamics and male–male bonding. Despite the high degree of arboreality of spider monkeys, all seven witnessed raids involved the males progressing single file on the ground in unusual silence. This is remarkably similar to the behavior of chimpanzees. The circumstances around the raids suggest that factors such as reduced mating opportunities, number of males relative to that in the neighboring community, and the strength of bonds among males could play a role in the timing of such actions. The raids did not appear to be aimed at finding food, whereas there is some indication that they may directly or indirectly increase reproductive opportunities. Although no killing was observed, the possibility cannot be excluded, according to the authors, that spider monkey raids may be aimed at harming rivals if a vulnerable individual were encountered. The similarity
68
JOHAN M. G. VAN DER DENNEN
of spider monkey raids with those of chimpanzees and humans supports the notion that lethal raiding is a convergent response to similar socioecological conditions. As there is no evidence for cooperative hunting in spider monkeys, the similarity of spider monkey raids with those of chimpanzees and humans could challenge the proposal of a link between group hunting and lethal raids based on the occurrence of both in humans and chimpanzees, but of neither in bonobos. (Aureli et al., 2006, pp. 494–495)
Bonobos Little is known yet about bonobo (P. paniscus) (also known as pygmy chimpanzee and gracile chimpanzee) intercommunity encounters. Encounters are mostly avoided, but when they occur they appear mildly antagonistic, ranging from peaceful mixing in the border area (and mainly confined to adult males giving branch-dragging displays), to nonlethal fighting (no observation has been made of participants killed in intergroup fights), and violent clashes sometimes leading to bloody wounds (Badrian & Badrian, 1984; Kano, 1984; Kano & Mulavwa, 1984; Knauft, 1991; Wrangham & Peterson, 1996). There appears to be intergroup dominance attenuating agonistic contacts (Kano, 1987; Kitamura, 1983; Nishida & HiraiwaHasegawa, 1987). Kuroda (1980) and Kano (1987) observed that male bonobos were the most aggressive sex during intercommunity encounters. The ‘‘make love, not war’’ species, the ‘‘ape from Venus’’ does not exist. In the wild, bonobos appear to be much less ‘‘sexy’’ and much more aggressive (including intergroup violence) than in semi-captivity (Vervaecke, 2002, p. 123). Even the first signs of hunting of black mangabeys by bonobos have been observed, as well as the consumption of monkey meat (Surbeck & Hohmann, 2008; Surbeck et al., 2009). Stanford (2001) pointed out that the contrast between bonobos and chimpanzees, as painted by de Waal (e.g., 2001) derives largely from a comparison of wild chimpanzees with captive bonobos. In bonobos, intercommunity encounters, too, often involve aggression, and within the group it is often the females who are ‘‘demonic’’ (Slurink, 2002, p. 272). Shea (1983) has argued that various bonobo morphological traits are neotenous and pedomorphic compared to common chimpanzees, and proposed that these changes resulted from selection for reduced sexual dimorphism in morphology and behavior. Wrangham and Pilbeam (2001, pp. 11–12) elaborated Shea’s idea with the specific suggestion that reduced sexual dimorphism functioned to reduce aggressive behavior by adult males.
The Biopolitics of Primates
69
Anatomically and behaviorally the bonobo (pygmy chimpanzee) may be characterized as a slender and pre-pubescent (pedomorphic or juvenilized) form of chimpanzee. Taxonomically speaking the two species P. troglodytes and P. paniscus should be renamed the robust and the gracile chimpanzees (Kortlandt, 1999, p. 25) (and humans the ‘‘third’’ chimpanzee as suggested by Diamond, 1992). Gracile chimpanzee society is a maternocracy. Another conspicuous characteristic of gracile chimpanzee society is its friendliness and peacefulness in combination with substantial male–female equality (e.g., de Waal & Lanting, 1998). Generally, young mammals are much more tolerant, amicable, and playful among themselves than adults. So it might be that the friendliness and peacefulness among the adult graciles is more or less an accidental byproduct of their anatomical and behavioral pedomorphism or juvenilization of the species (Kortlandt, 1999, pp. 35–36). Kano (1987) made the intriguing suggestion that in the pygmy chimpanzees the ‘‘in-group feeling’’ among females is very strong, and therefore aggressive male expansion of territory is not connected with an increase in available females, and thus does not payoff. Even contrasts between the male-dominated hierarchies of chimpanzees, on the one hand, and the more egalitarian relationships of bonobos, on the other hand, come down to the degree to which individuals or groups can monopolize resources, such as food and mates, that are important to reproduction (Strier, 2001, pp. 78–79).
THE EXTENT OF ANIMAL INTERGROUP AGONISTIC BEHAVIOR Collective defensive behavior is highly differential among animal species. Most of them have neither defense nor offense in their repertoire. Collective defense is exemplified by avian mobbing of raptors, and collective territorial defense in some other bird species. In some avian species territorial defense involves regular patrolling of territorial boundaries and chorus vocalizations by all group members in concert. Among social carnivores, a number of species show coordinated lethal attacks. In wolves (Canis lupus), family-based packs occasionally invade neighboring packs’ territories, attacking residents; Mech (1977) found that intraspecific conflict accounted for 43% of deaths not caused by humans. Among spotted hyenas (Crocuta crocuta), who, like wolves, live in
70
JOHAN M. G. VAN DER DENNEN
family-based, territory-holding groups, intruders into a clan’s territory are likely to be attacked and killed, and smaller clan subgroups patrol the territory boundaries, confronting other ‘‘patrols’’ (Kruuk, 1972, 1975). Neighboring clans sometimes engage in pitched battles over carcasses of prey that one or the other of the groups has killed. In lions (Panthera leo), which also live in groups (prides) based on a group of related females and one or more associated males, interpride encounters occur, but lethal injury is rare. When invading males are attempting to take over a pride, there may be lethal injuries, though once one male cedes reproductive rights, aggression typically stops. New males are likely to commit infanticide (Bertram, 1976, 1978; Packer & Pusey, 1982–1984). Collective defense would be rather pointless if there were no threat or danger of offense. In the remainder of this chapter I shall focus on these (mainly primate) species in which intraspecific IAB, involving observed offensive episodes, has been clearly ascertained.
General Observations on Intergroup Agonistic Behavior 1. The overwhelming majority of gregarious and social mammalian species does not have IAB in its behavioral repertoire. It is likely that these species lack the requisite social and cognitive skills, such as a coalitional psychology, to cooperate polyadically for the sake of orchestrated intergroup competition: a high-risk/high-gain strategy. 2. The majority of species in which IAB has been documented belong to the primate order. The intergroup behavior of primates is extremely variable – in terms of frequency and intensity of encounters (even between different populations of the same species), the resources being contested, and the sex of the participants – and ranges from very relaxed and ‘‘peaceful’’ to lethal raiding (Brown, 1975; Eberhart & Candland, 1981; Cheney, 1987; Crofoot & Wrangham, 2009). Some level of intergroup aggression occurs during intergroup encounters in most primates whether or not they actually defend a home range (Fashing, 2001, p. 219). Escalated aggression rarely occurs during intergroup conflicts in groupliving primates. One possible explanation is that the costs of aggression, compared to the potential benefits, are high (Crofoot & Wrangham, 2009, p. 18). An alternative explanation is that the collective action problem inherent in group-level resource competition presents an obstacle to high individual investment (Nunn & Lewis, 2001; Nunn & Deaner, 2004;
The Biopolitics of Primates
71
Kitchen & Beehner, 2007; Crofoot & Wrangham, 2009, pp. 18–19), though protecting reproductive investment seems, at least partly, to explain the willingness of individuals to defend the group, even when others do not assist (Kitchen & Beehner, 2007, pp. 1566–1567). Also dominance rank explains behavioral variation in agonistic intergroup participation in several species [capuchins (Cebus capucinus): Fragaszy, Visalberghi, & Fedigan, 2004; Tibetan macaques (Macaca thibetana): Zhao, 1997; blue monkeys: Payne, Lawes, & Henzi, 2003; black howlers (Alouatta pigra): Kitchen, Cheney, & Seyfarth, 2004; ring-tailed lemurs: Nunn & Deaner, 2004; bonnet macaques (Macaca radiata): Cooper, Aureli, & Singh, 2004]. Thus, collective action problems are solved in several species because the animals reaping the benefits are the ones providing the service (Kitchen & Beehner, 2007, pp. 1564–1565). In most primate species conflicts between groups are rare. Neighboring groups generally avoid each other. When groups spot one other, the group most eccentric in relation to its home range generally is the first to retreat, or, in case a dominance–subordination relationship exists among the groups (vide infra), the subordinate group retreats (Kawanaka, 1973; Eberhart & Candland, 1981; McKenna, 1982; Cheney, 1983, 1987). Among primates, exclusive use of space is generally maintained by (a) site attachment and avoidance of the ranges of neighboring groups (mutual proximity-dependent avoidance); (b) site-dependent aggression and regular definition of the conventional location of boundaries; and (c) active defense of (exclusive access to) an area’s resources by advertisement and/or eviction of intruders (territoriality). The behavioral mechanisms regulating spacing and grouping may vary within a genus and even within a species. For example, in a Callicebus torquatus (yellow-handed titi monkey) population in Peru, exclusive use of space is maintained by mutual avoidance and restricting movements to familiar areas (Kinzey & Robinson, 1983). In contrast, Callicebus moloch (dusky titi) has been described as ‘‘territorial’’ (Mason, 1966, 1968) in that, in addition to groups occupying exclusive areas, spacing between groups is maintained by site-dependent aggression: the probability that a group will attack, rather than avoid, another group depends on the site at which the encounter takes place. In this (and other) species, that probability is low at the center of the group’s own range, increases the closer the group is to the boundary, and then drops off rapidly as the boundary is crossed. The outcome of an aggressive encounter therefore varies with locality. Each group is more aggressive and therefore displaces other groups more easily when it is within its own exclusive area. Groups are most aggressive close to, but on their own side of, the boundary, a ‘‘doughnut’’-shaped aggression field (Waser & Wiley, 1979) that results in the
72
JOHAN M. G. VAN DER DENNEN
clear definition and reinforcement of the conventional location of the boundaries. Typically, chasing occurs, but physical contact is rare (Robinson, 1979, 1981). The ordinal positions of dominance among groups are thus contingent upon their positions on a central–peripheral axis across their territories (Carpenter, 1974). Groups of ring-tailed lemurs, for example, win interactions in their ‘‘typical’’ ranges and tend to lose outside these areas, regardless of the strength of the opposing group (Pride, Felantsoa, Randriamboavonjy, & Randriambelona 2006). Intergroup confrontations between dusky titi appear to be very frequent, conspicuous, regular, and predictable, occurring at stable boundary areas, and either limited or markedly affected by time of day (Klein, 1974). When two or more groups of primates meet, the resulting behavior may range from complete fusion of the groups (no agonism), via threats and displays, fights and chases, to outright killing. Typical outcomes of primate group encounters in escalating order of antagonism (Eberhart & Candland, 1981) are given in Table 1. As a rule, most primate agonistic group encounters are of the ‘‘ritualized contact’’ type, in which injuries are rare and hardly serious, and fatalities virtually unknown (Eberhart & Candland, 1981; King, 1980). The few times when contact between groups results in a collective skirmish, it seems to be the result of incidental escalation of brawls between peripheral individuals in which others interfere, rather than a deliberately coordinated and concerted enterprise (van Hooff, 1990a). Some nonterritorial species have evolved loud calls (‘‘duetting calls’’) that aid in the regulation of intergroup spacing and mutual avoidance. In contrast, when groups defend all or part of their home ranges, most intergroup interactions are characterized by aggression rather than by mutual avoidance. Friendly interactions (play, grooming, copulation) between members of different intraspecific groups do, however, occur in both territorial and nonterritorial species (Cheney, 1987, p. 272). The agonistic repertoire in these encounters ranges from vocal and gestural displays, bluffing and intimidation, through threats, chases, ‘‘pitched battles’’ to ‘‘deliberate’’ killing. The level of agonism in macaques depends on the previous history of the troops, familiarity of the members, rank of the troops, and location of the encounter (Chapais, 1983; Ciani, 1986; Hausfater, 1972; Vessey, 1968). 3. Old World monkeys show various types of spatial and social organization, but in most cases they use variants of the group home range system, with sometimes as much as 80% overlap between groups (Hall, 1965). Communal defense of home range or territory against intraspecific
73
The Biopolitics of Primates
Table 1.
Outcomes of Primate Group Encounters (After Eberhart & Candland, 1981).
Description Complete fusion Fission–fusion Indifference Mutual avoidance
Unilateral avoidance
Mild threat
Intensive threat Ritualized contact
Injurious contact Killing
Characteristic Groups coalesce permanently, with restructuring of social relations (e.g., rhesus monkeys: Bernstein et al., 1974) Groups aggregate and interact, but subsequently diverge (e.g., mountain gorilla: Schaller, 1963) Groups in proximity, but appear indifferent to, or ignore each other (e.g., Himalayan langurs: Sugiyama, 1976) Coordination of range use such that groups seem to avoid contact (e.g., prosimians: Sussman & Richard, 1974; owl monkeys: Wright, 1978) Avoidance or displacement of one group by another (e.g., baboons: Nash, 1976; rhesus: Gabow, 1973; Japanese macaques: Kawanaka, 1973) Groups engage in gestural, locomotor, or vocal displays or threats (e.g., squirrel monkeys: Baldwin & Baldwin, 1976; bonobo: Nishida & Hiraiwa-Hasegawa, 1987) High intensity threats involving chases, physical contact between animals is rare (e.g., Kloss’s gibbon: Tenaza, 1975) Aggressive physical contact between groups, but serious injury occurs rarely (e.g., black-and-white colobus: Schenkel & Schenkel-Hilliger, 1967) Severe aggressive contact, with injury or death in both groups (e.g., cercopithecines: Struhsaker, 1969) Death of one or more animals in one group (e.g., squirrel monkeys: Candland et al., 1978)
intruders seems to be the most common manifestation of IAB in primates and carnivores. However, relatively few primate species maintain true territories in the sense of defended spatial exclusiveness (Crook, 1968; Rowell, 1972; Schuster, 1978; Scott, 1969; Vine, 1973). Rather, there is some overlapping of foraging ranges in most studied species of Cercopithecoidea and Ceboidea. According to Jolly (1972), the cercopithecoids (macaques and baboons) as a group are by far the most aggressive among the primates. They also seem to have the most thoroughly organized societies (Nagel & Kummer, 1974). The guenonpatas group, which ranges from forest to savanna, tends to have a low level of intragroup aggression in either place. Troops of rhesus macaques (Southwick, 1962), Japanese macaques (Kawanaka, 1973), langurs (Sugiyama, Yoshiba, & Parthasarathy, 1965;
74
JOHAN M. G. VAN DER DENNEN
Ripley, 1967), and howlers (Southwick, 1962, 1963; Carpenter, 1965), avoid contact with each other when approaching the overlapping zones. When two howler troops approach each other, they will engage in a noisy howling battle, but Southwick (1963) reported that he never saw two howlers fight. Adjacent social groups in the majority of colobines generally interact aggressively (Struhsaker & Leland-Struhsaker, 1987, p. 91; Yeager & Kirkpatrick, 1998), with adult males usually the most aggressive and frequent participants. In most populations of Presbytis entellus, adult females (Ripley, 1967), as well as youngsters (Hrdy, 1977a, 1977b) may play prominent roles in intergroup fights. In at least one population of this species, however, neighboring groups were extremely tolerant of one another (Jay, 1965). Poirier (1974) summarized the major characteristics of colobine aggression (genera Colobus and Presbytis). Much of colobine aggressive behavior, and certainly some of its most dramatic instances, are witnessed between rather than within troops. Among some colobines, extreme aggression is witnessed between males of bisexual and all-male groups. When members of the latter invade a bisexual troop, infanticide and group leadership change is common. Oates (1977) studied the black-and-white colobus of the Kibale Forest, West Uganda. Relationships between guereza groups were generally hostile, and in some habitats groups defended nearexclusive territories. While the Kibale groups had extensive range overlap, they still interacted aggressively in the majority of encounters. Poirier (1970) reported about the intergroup hostility in Nilgiri langurs [Presbytis 9Trachypithecus0 johnii]. Adult male Nilgiri langurs regularly sought conditions for displaying against adult males of other troops. An adult male frequently deserted his troop to challenge another adult male. This is striking because Nilgiri langurs possess means of signaling troop locations and avoidance that could substitute for adult male aggression in the spacing of troops. Home range size is influenced by both dietary requirements and the spatiotemporal distribution of food (Cheney, 1987; Clutton-Brock & Harvey, 1977). A group’s ability to patrol its range on a regular basis is correlated with the presence or absence of territoriality (Mitani & Rodman, 1979). This is in accordance with the ‘‘economic defendability’’ model of territoriality (Brown, 1964, 1975). 4. Facilitating and/or aggravating conditions of IAB in primates include: extreme crowding and population density due to, for example, areal reduction, competition for food at artificial feeding sites, and other rather
The Biopolitics of Primates
75
‘‘pathological’’ conditions such as disruption caused by human interference. Probably the most fierce intertroop fighting observed in monkeys was that described by Southwick et al. (1965) among rhesus monkeys living in a temple in India where there was 90% overlap between the ranges of three troops. Severe fighting was almost a daily occurrence and all the males bore scars and often sustained broken bones. While adult males usually started the fight, females and juveniles joined in. 5. In primate species which form all-male groups, defense by the leader of the bisexual group against the violent appropriation and abduction of females by the all-male group may result in prolonged and severe fighting with skirmishes and raids lasting for periods of 3–10 days (Bygott, 1979; Mohnot, 1971). In capped langurs [Presbytis pileata (Trachypithecus pileatus)], intergroup aggression is not related to defense of food sources by either sex, but rather appears to involve attempts by males from outside the group to interact with group females. During intergroup encounters, resident females bite and push females that approach an intruding male. Males appear to use intergroup encounters as a means of defending their own females while gaining access to those of other groups (Stanford, 1991). Mate defense and the exploration of new breeding opportunities appear to be important functions of intergroup conflict also in moustached tamarins (Saguinus mystax) (Garber, Prietz, & Isaacson, 1993) and baboons (Papio spp.) (vide supra). 6. For some primate species agonistic intertroop or intercommunity encounters may be highly attractive (at least for some individuals, mostly young males) (chimpanzees: Bygott, 1979; Goodall, 1986; gibbons: Ellefson, 1968; mountain gorillas: Fossey, 1979; langurs: Ripley, 1967; Morrison & Menzel, 1972), and confrontations may be actively sought and provoked (which suggests intrinsic motivation: They seem to be ‘‘spoiling for a fight’’). On the other hand, it has become increasingly clear lately that female involvement in IAB has been systematically underestimated. For example, among rhesus macaques at Cayo Santiago, ‘‘violent intergroup squabbles y were marked by sustained fighting in line formation y 2–20 animals faced off with individuals of an opposing group and reciprocally lunged, batted, and growled y participants in the line were most often adult females and 2- to 5-year-old males (i.e., juveniles and subadults)’’ (Hausfater, 1972). Adult females may be more
76
JOHAN M. G. VAN DER DENNEN
involved than males and in a number of species are reported to constitute the central phalanx (Manson & Wrangham, 1991). In many territorial primate species, females are frequently aggressive during intergroup interactions [e.g., ring-tailed lemurs, emperor (Saguinus imperator) and saddleback tamarins (Saguinus fuscicollis), vervets, redtail monkeys, blue monkeys, and Kloss’s gibbons (Hylobates klossii)]. Female aggression is more variable, however, in species that only infrequently defend ranges. In some of these, female aggression is common (e.g., macaque spp., red howlers, capuchins). In other species, however, males are the primary antagonists, perhaps because, as Cheney (1987) suggested, encounters more often concern mate, rather than food, defense (e.g., baboons, mountain gorillas). The wild mountain gorilla (Gorilla gorilla berengei) is organized into agegraded-male troops. Dominance behavior is very low-keyed and overt aggression nearly nonexistent (Schaller, 1963, 1965; Fossey, 1971). Groups respond in variable and unpredictable ways when they meet. Usually the encounters are peaceful, but mutual aggression and aversion, and sometimes severe violence, also occur on occasion. Fossey has stressed the importance of the personal idiosyncracies of the dominant males (silverbacks), who control the movements of the group. In primate species characterized by male dispersal (or female philopatry), female hostility toward other groups and cooperation in intergroup aggression is common, and may involve both resource defense against extra-group females, and, more or less collective, antagonism toward migrant, potentially infanticidal, males. Gang attacks, involving mostly females, have twice been reported to lead to deaths of males attempting to enter a group of red colobus (Manson & Wrangham, 1991). Cheney (1987, p. 278) noted that the hostility of the females toward intruder males often escalates and affects the whole group. Female antagonism toward extragroup females also occurs in some of the social carnivores. In monogamous primate species, such as the gibbons (Hylobates spp.) and their close relative the siamang (Symphalangus syndactylus), females may be as aggressively participating in cooperative range or territory defense and other intergroup interactions as males. In these situations the animals tend to be most aggressive toward individuals of their own sex, perhaps because they represent potential mate competition (Cheney, 1987, p. 275). Gibbons maintain their family territories largely by vocalizations and bluffing behavior, which prevents a great deal of actual fighting. Despite this, Carpenter (1940) noted scars of healed wounds on many animals that
The Biopolitics of Primates
77
indicate serious fighting does occur. Monogamy in the gibbons is enforced by hostility between the females, even between mother and daughter (Leighton, 1987; Cheney, 1987, p. 275). In primate species characterized by female dispersion, in which the females transfer to new groups, on the other hand, females tend not to participate in aggressive intergroup interactions: for example, in gorillas (Fossey, 1979; Harcourt, 1978), red colobus (Struhsaker & LelandStruhsaker, 1979), and hamadryas baboons (Abegglen, 1984). Among chimpanzee females attacks on females of other communities are rare, but not entirely absent (Goodall et al., 1979; Cheney, 1987). Female involvement in intergroup aggression has been proposed to depend on whether resources that limit female reproduction are defensible through cooperative action. Accordingly, species in which females cooperate may be expected to be those with female philopatry. Female philopatry also appears to be associated with female participation in intergroup aggression in social carnivores such as spotted hyenas (Kruuk, 1972) and lions (Packer, Scheel, & Pusey, 1990). 7. In a number of primate and carnivore species IAB is accompanied by infanticide and, occasionally, cannibalism. For the evolutionary rationale behind infanticide, as a male reproductive strategy, see Hausfater and Hrdy (1984) who also present evidence on other genera and taxa, and van Schaik and Janson (2000). 8. Personal idiosyncrasies and ‘‘character structure’’ of (mostly male) leaders strongly influence group integration, the movements of the group within its home range, and dominance–subordinate relationships with other groups (Itani, Tokuda, Furuya, Kano, & Shin, 1963; Kawai, 1964; Fossey, 1971, et seq.; Kawanaka, 1973). Careful control of trouble within the group and leading attacks on other groups was for instance characteristic of some leader males at Takasakiyama. Furthermore, concerted action and scouting behavior of group males seem to indicate a ‘‘consciousness of belonging’’ (Kawanaka, 1973) in this species. In the Mahale Mountains area, inter-unit-group antagonism is based mainly on the antagonism among adult chimpanzee males who belong to different unit-groups (Nishida, 1979). Furthermore, individuals may vary in their willingness to participate in intergroup aggression based on characteristics such as their age and experience (e.g., Lazaro-Perea, 2001), fighting ability (e.g., Steenbeek, 1999), changing state (e.g., lactating: Lazaro-Perea, 2001), and temperament (Kitchen & Beehner, 2007, p. 1573).
78
JOHAN M. G. VAN DER DENNEN
9. When home ranges overlap extensively, the aggressive defense of a particular resource may be more costly than the simple avoidance of other groups. In such cases, intergroup competition is often mediated by the relative dominance of the groups involved. There is evidence for a positive correlation between a group’s size (and the number of adult males) and its ability to displace other groups (though occasionally more subtle factors – such as the history of past relations between the groups – are involved) (Brown, 1975). Observational studies support the expectation that larger groups should tend to win intergroup fights (Cheney, 1987; Packer et al., 1990; Harcourt & de Waal, 1992; Gat, 1999; Wrangham, 1999; Boesch & Boesch-Achermann, 1999/2000; Wilson et al., 2002; Thayer, 2004; Crofoot & Wrangham, 2009). This results in a definite linear group dominance hierarchy revealed by approach–retreat encounters. Approach–retreat encounters have, for example, been observed in most macaque species (Givens, 1975; Saito et al., 1998; Crofoot & Wrangham, 2009). Occasionally, groups expand their home ranges at the expense of their neighbors’, and in these cases the relative sizes of the groups – as well as the fighting ability of a particular individual – may determine success (Cheney, 1987, pp. 271–272). Robinson (1988) demonstrated a linear dominance hierarchy among groups of wedge-capped capuchins (Cebus olivaceus). A group’s overall ability to dominate another may influence the behavior of its members (Kitchen & Beehner, 2007, p. 1568). 10. With regard to the general function(s) of IAB in primates, Washburn and Hamburg (1968) succinctly epitomized: ‘‘Intergroup aggression either leads to one group’s having the resources of an area at its exclusive disposal, or at least creates a situation in which one is much more likely to obtain food in one area.’’ Food is not the only resource nonhuman species fight over, however. Much more important is the struggle for differential reproduction. Nonhuman vertebrate males frequently come into open conflict over access to females, and/or control of resources useful in attracting females. Females, on the other hand, may cooperate in coalitions of kin to attack reproductive competitors, or the offspring of reproductive competitors (Wasser, 1983; Silk & Boyd, 1983); such situations typically involve harassment of subordinate females and infanticide, with little risk to the aggressors (see review by van der Dennen, 1992). The basic reasons for male–male intergroup aggression, rather than intergroup aggression by both sexes,
The Biopolitics of Primates
79
probably include the different reproductive payoff curves for the two sexes in mammals generally (Low, 1993). As Alexander (1989), and G. Johnson have emphasized: ‘‘Competition is often attributed to scarcity, but to understand fully the ubiquity in competition in a Darwinian world, it must be understood that competition normally occurs even in the absence of scarcity y there will be competition to acquire unequal proportions y In addition, certain reproductively relevant resources are inherently scarce’’ (G. Johnson, 1995, pp. 293–294). In other words, in a Darwinian world, it is impossible not to compete.
EXPLANATIONS OF LETHAL RAIDING IN CHIMPANZEES Goodall (1986, pp. 526–533) herself explained the chimpanzee protowarfare in terms of the idiosyncratic pattern of chimpanzee territoriality and preadaptations common in chimpanzees and early humans. In three important ways, she explains, chimpanzee behavior does not comply with classical territoriality: (a) both at Gombe and Mahale it is the relative size and the composition of the two neighboring parties that determine the outcome of an encounter, rather than the geographic location; (b) chimpanzees have a large home range with considerable overlap between neighboring communities; and (c) it is perhaps in the violence of their hostility toward neighbors that chimpanzees, like hyenas and lions, differ most from the traditional territory owners of the animal kingdom. Their victims are not simply chased out of the owners’ territory if they are found trespassing; they are assaulted and left, perhaps to die. Moreover, chimpanzees not only attack trespassers, but may make aggressive raids into the very heart of the core area of neighboring groups. It is also of considerable interest to find that the chimpanzees show behaviors that bear strong resemblance to, and hence may be precursors to pseudospeciation in humans. First, their sense of group identity is strong; they clearly differentiate between in-group and out-group, between individuals who ‘‘belong to us’’ and those who do not. This sense of group identity is, Goodall claims, far more sophisticated than mere xenophobia. The members of the Kahame chimpanzee community had, before they split, enjoyed close and friendly relations with their aggressors. By separating themselves, it is as though they forfeited their ‘‘right’’ to be treated as group members – instead they were treated as strangers. Second, nongroup
80
JOHAN M. G. VAN DER DENNEN
members may not only be violently attacked, but the patterns of attack may actually differ from those utilized in typical intracommunity aggression. The victims are treated more as though they were prey animals; they are ‘‘dechimpized.’’ Finally, Goodall observed, chimpanzees appear to possess the cognitive sophistication which is a prerequisite for the genesis of cruelty: they are capable to some extent of imputing desires and feelings to others, and they are almost certainly capable of feelings akin to (human) sympathy and empathy. Several hypotheses have been put forward to explain the occurrence of lethal raids in chimpanzees, both in terms of the proximate mechanisms and the long-term functions (e.g., Wrangham, 1979, 1999; Wilson & Wrangham, 2003). They can be reduced to two main hypotheses (Aureli et al., 2006). Lethal raids can be viewed as an extreme form of mate competition in which males kill rival males to defend access to their own females, and possibly to gain access to females from neighboring communities [the ‘‘maleonly community hypothesis’’: Wrangham, 1979; Manson & Wrangham, 1991; Boesch & Boesch-Achermann, 1999/2000; theoretically, according to this view, the ultimate benefit of intergroup aggression among chimpanzees is expected to be increased access by aggressive males to reproductively valuable females, via either incorporation of neighbors or encroachment on the territory of neighboring males (Manson & Wrangham, 1991; cf. Low, 1993; van der Dennen, 1995)]. Another possibility is that lethal raids are an extreme form of intercommunity feeding competition in which males defend or expand their territory, thus increasing the availability of food sources for themselves, the females of their community, and their offspring [the ‘‘bisexual community hypothesis’’ (implicit in the original ideas of Nishida and colleagues): Wrangham, 1999; Pusey, 2001; Pusey, Williams, & Goodall, 1997; Williams, Oehlert, & Pusey, 2004]. Since Wrangham formulated this idea of the male-only community hypothesis, evidence against it has accumulated. First, in addition to showing aggression to strange males, when males at the edges of the territory encounter stranger females who are not sexually receptive, they often attack them severely – even lethally. Goodall described severe attacks between 1971 and 1982 on 20 different stranger females at Gombe, 15 of whom had infants or juveniles. In three attacks the males killed the females’ infants, but in most cases the aggression was directed primarily at the female. In 13 of 14 cases in which the female was well observed, she incurred serious wounds or left a great deal of blood. In no case were the males observed associating with these females after the attack (Pusey, 2001, p. 18;
The Biopolitics of Primates
81
Wrangham, 2006). Second, females are sensitive to the position of the community border. What then is the purpose of a male territoriality whereby males repel not only males but also females of other communities? Pusey et al. (1997) found evidence that the males are defending a feeding territory for all the community members. By expanding the community range and thereby excluding female as well as male feeding competitors, the males increased the reproductive rates of the resident females. If the community range remains large, the number of females may eventually increase as dispersing females settle in the area, but females do not seem to be ‘‘captured’’ in the course of the initial expansion as suggested by the male-only community hypothesis (Pusey, 2001, p. 19).
The Imbalance-of-Power and Intergroup Dominance Hypotheses Why should chimpanzees, but few other mammals, engage in lethal intergroup attacks? The imbalance-of-power hypothesis developed by Wrangham and colleagues (Manson & Wrangham, 1991; Wrangham & Peterson, 1996; Wrangham, 1999, 2001; Crofoot & Wrangham, 2009) attempts to answer this question by drawing attention to the costs, rather than the benefits, of aggression. A growing number of studies support the view that animals use aggression strategically, when the benefits are likely to outweigh the costs of aggression (Archer, 1988; Huntingford & Turner, 1987; Wilson, 2003; Wilson & Wrangham, 2003). The imbalance-of-power hypothesis proposes that in chimpanzees fission– fusion social structure, combined with coalitionary bonds among males, creates opportunities for low-cost killing of rivals. Chimpanzees travel in parties of varying size and composition, which may result in large disparities in party size during intergroup encounters, even among communities that are similar in overall size. In such cases, members of the larger group can kill rivals at very low risk of injury to themselves. During observed lethal intergroup attacks, three or more attackers ganged up on a single victim, who was immobilized by some of the attackers while others beat and bit the victim. Attackers themselves rarely received injuries. The imbalance-ofpower hypothesis thus argues that intergroup attacks are lethal not because of unusually high benefits to be obtained from killing, but because the costs of killing are low during gang attacks. The primary, and ultimate, benefit of intergroup killing is thought to be the reduction of the coalitionary strength of rival communities, or, in other words, the weakening of the territorial
82
JOHAN M. G. VAN DER DENNEN
power of the neighboring communities (Wrangham, 2001, p. 265). By reducing the fighting strength of a given community, males increase their chances of success in future battles with that community. More frequent success in battles should result in territory expansion and thus more food for group members, including females and offspring (Wrangham, 1999, pp. 11–12; Wilson & Wrangham, 2003, p. 381). It pays for groups to achieve dominance over neighboring groups so that they can enlarge their territories. Also Wilson, Wallauer, and Pusey (2004) research supports the view that intergroup violence is a persistent feature of chimpanzee societies and that the primary benefit attackers gain from them is reduced competition for resources. Furthermore, they direct their attacks almost entirely to males, the sex that alone defends the territory. Thus while female strangers are often found alone and may be chased and beaten, more than 90% of adult victims of lethal intercommunity aggression are male. Even infants killed by adults are mostly male. Those who die, therefore, are mainly the present and future defenders of the territory (Wrangham, 2006). In the event of a successful attack there is no immediate payoff other than the satisfaction the aggressors experience from the act itself. The implication is that natural selection has favored in chimpanzees a tendency to relish the prospect and performance of such brutality (Wrangham, 2006). The idea that the low cost of lethal aggression elicits lethal raiding is central to the imbalance-of-power hypothesis. The imbalance-of-power hypothesis also predicts that the aggressors will be members of the philopatric sex, whether females (as in spotted hyenas, Crocuta crocuta) or males (as in chimpanzees) (Wrangham, 1999; Crofoot & Wrangham, 2009, p. 22). Chimpanzees may take large risks when potential benefits are large or when failure to do so could inflict larger costs. Males in communities with a relatively small number of sexually active females might take more risks to attack groups with more females to try and improve their reproductive success, even if those groups have many more males (Boesch et al., 2008). Recently, Mitani, Watts, & Amsler (2010) found evidence that chimpanzees at Ngogo attack neighbors to expand their territories and to gain access to more food. The researchers concluded that a causal link between lethal intergroup aggression and territorial expansion can be made now that the Ngogo chimpanzees use the area once occupied by some of their victims.
83
The Biopolitics of Primates
REFERENCES Abegglen, J. J. (1984). On socialization in Hamadryas baboons. Cranbury, NJ: Associated University Presses. Adang, O. (1999). De machtigste chimpansee van Nederland: leven en dood in een mensapengemeenschap. Amsterdam: Nieuwezijds. Albrecht, H., & Dunnett, S. C. (1971). Chimpanzees in Western Africa. Munich: Piper. Alexander, G. M., & Hines, M. (2002). Sex differences in response to children’s toys in nonhuman primates (Cercopithecus aethiops sabaeus). Evolution and Human Behavior, 23(6), 467–479. Alexander, R. D. (1989). Evolution of the human psyche. In: P. Mellars & C. Stringer (Eds), The human revolution: Behavioral and biological perspectives on the origins of modern humans (pp. 455–513). Princeton, NJ: Princeton University Press. Altmann, S. A. (1956). Avian mobbing behavior and predator recognition. Condor, 58(4), 241–253. Altmann, S. A. (1962). A field study of the sociobiology of rhesus monkeys (Macaca mulatta). Annals, 102, 338–435. Archer, J. (1988). The behavioural biology of aggression. Cambridge, UK: Cambridge University Press. Arnold, K., & Aureli, F. (2006). Postconflict reconciliation. In: C. J. Campbell, A. Fuentes, K. C. McKinnon, M. Panger & S. K. Bearder (Eds), Primates in perspective (pp, 592–608). Oxford: Oxford University Press. Aureli, F., Schaffner, C.M., Verpooten, J., Slater, K., & Ramos-Fernandez, G. (2006). Raiding parties of male spider monkeys: Insights into human warfare? American Journal of Physical Anthropology, 135(4), 486–497. Badrian, A., & Badrian, N. (1984). Social organization of Pan paniscus in the Lomako Forest, Zaire. In: R. L. Susman (Ed.), The pygmy chimpanzee: Evolutionary biology and behavior (pp. 325–346). New York: Plenum Press. Baldwin, J. D., & Baldwin, J. I. (1976). The vocalizations of howler monkeys (Alouatta palliata) in southwestern Panama. Folia Primatologica, 26, 81–108. Baldwin, J. D., & Baldwin, J. I. (1981). The squirrel monkeys, genus Saimiri. In: A. F. CoimbraFilho & R. A. Mittermeier (Eds), Ecology and behavior of neotropical primates (Vol. 1). Rio de Janeiro: Academia Brasileira de Ciencias. Bernstein, I. S. (1999). Animal behavior studies: Primates. In: L. R. Kurtz (Ed.), Encyclopedia of violence, peace, and conflict (Vol. 1, pp. 67–76). San Diego, CA: Academic Press. Bernstein, I. S., Gordon, T. P., & Rose, R. M. (1974). Factors influencing the expression of aggression during introductions to rhesus monkey groups. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 21–40). New York: Academic Press. Bernstein, I. S., & Ehardt, C. L. (1985). Intragroup agonistic behavior in rhesus monkeys (Macaca mulatta). International Journal of Primatology, 6(3), 209–226. Bertram, B. C. R. (1976). Kin selection in lions and evolution. In: P. P. G. Bateson & R. A. Hinde (Eds), Growing points in ethology (pp. 281–301). Cambridge: Cambridge University Press. Bertram, B. C. R. (1978). Pride of lions. New York: Scribner’s. Boesch, C., & Boesch-Achermann, H. (1999/2000). The chimpanzees of the Taı¨ Forest: Behavioral ecology and evolution. New York: Oxford University Press.
84
JOHAN M. G. VAN DER DENNEN
Boesch, C., et al. (2008). Intergroup conflicts among chimpanzees in Taı¨ National Park: Lethal violence and the female perspective. American Journal of Primatology, 70, 519–532. Brown, J. L. (1964). The evolution of diversity in avian territorial systems. Wilson Bulletin, 76, 160–169. Brown, J. L. (1975). The evolution of behavior. New York: Norton. Bygott, J. D. (1972). Cannibalism among wild chimpanzees (Pan troglodytes). Nature, 238, 410–411. Bygott, J. D. (1979). Agonistic behavior, dominance, and social structure in wild chimpanzees of the Gombe National Park. In: D. A. Hamburg & E. R. McCown (Eds), The great apes: Perspectives on human evolution (Vol. 5, pp. 405–428). Menlo Park, CA: Benjamin/ Cummings. Bygott, J. D. (1974). Agonistic behaviour and social relationships among adult male chimpanzees. Ph.D. Thesis, University of Cambridge. Candland, D. K., et al. (1978). Behavior of unacquainted Saimiri troops upon encounter: A suggestive case study. Primates, 19(4), 643–655. Carpenter, C. R. (1940). A field study in Siam of the behavior and social relations of the gibbon (Hylobates lar). Comparative Psychological Monographs, 16(5), 1–212. Carpenter, C. R. (1965). The howlers of Barro Colorado Island. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 250–291). New York: Treubner King. Carpenter, C. R. (1974). Aggressive behavioral systems. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 459–496). New York: Academic Press. Chapais, B. (1983). Autonomous, bisexual subgroups in a troop of rhesus monkeys. In: R.A. Hinde (Ed.), Primate social relationships: An integrated approach (pp. 220–222). London: Blackwell. Chapais, B. (1991). Primates and the origin of aggression, power, and politics among humans. In: J. Loy & B. Peters (Eds), Understanding behavior: What primate studies tell us about human behavior (pp. 190–228). New York: Oxford University Press. Cheney, D. L. (1981). Intergroup encounters among free-ranging Vervet monkeys. Folia Primatologica, 35, 124–146. Cheney, D. L. (1983). Intergroup encounters among Old World monkeys. In: R. A. Hinde (Ed.), Primate social relationships: An integrated approach (pp. 233–241). London: Blackwell. Cheney, D. L. (1987). Interactions and relationships between groups. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 267–281). Chicago: University of Chicago Press. Cheney, D. L. (1992). Intragroup cohesion and intergroup hostility: The relation between grooming distributions and intergroup competition among female primates. Behavioral Ecology and Sociobiology, 3(4), 334–345. Ciani, A. C. (1986). Intertroop agonistic behavior of a feral rhesus macaque troop ranging in town and forest areas in India. Aggressive Behavior, 12(6), 433–439. Clutton-Brock, T. H., & Harvey, P. H. (1977). Primate ecology and social organization. Journal of Zoology, 183, 1–39. Clutton-Brock, T. H., & Parker, G. A. (1995). Sexual coercion in animal societies. Animal Behavior, 49(5), 1345–1365. Collins, D. A. (1981). Social behaviour and patterns of mating among adult yellow baboons (Papio c. cynocephalus L. 1766). Ph.D dissertation, University of Edinburgh.
The Biopolitics of Primates
85
Cooper, M. A., Aureli, F., & Singh, M. (2004). Between-group encounters among bonnet macaques (Macaca radiata). Behavioral Ecology and Sociobiology, 56, 217–227. Cords, M. (1984). Mating patterns and social structure in redtail monkeys (Cercopithecus € Tierpsychologie, 64(3–4), 313–329. ascanius). Zeitschrift fur Cords, M. (1987). Forest guenons and patas monkeys: Male-male competition in one-male groups. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 98–111). Chicago: University of Chicago Press. Crofoot, M. C., & Wrangham, R. W. (2009). Intergroup aggression in primates and humans: The case for a unified theory. In: P. M. Kappeler & J. Silk (Eds), Mind the gap: Tracing the origins of human universals (pp. 171–197). Berlin: Springer. Crook, J. H. (1968). The nature and function of territorial aggression. In: M. F. A. Montagu (Ed.), Man and aggression (pp. 141–178). London: Oxford University Press. Daniel, J. R., Santos, A. J., & Cruz, M. G. (2009). Postconflict behaviour in brown capuchin monkeys (Cebus apella). Folia Primatologica, 80(5), 329–340. de Waal, F. B. M. (1978a). Exploitative and familiarity-dependent support strategies in a colony of semi-free living chimpanzees. Behaviour, 66(3–4), 268–312. de Waal, F. B. M. (1978b). Join-aggression and protective aggression among captive Macaca fascicularis. In: D. J. Chivers & J. Herbert (Eds), Recent advances in primatology, Vol. 1: Behaviour (pp. 577–579). New York: Academic Press. de Waal, F. B. M. (1982). Chimpanzee politics: Power and sex among apes. London: Jonathan Cape. de Waal, F. B. M. (1984). Sex differences in the formation of coalitions among chimpanzees. Ethology and Sociobiology, 5(4), 239–255. de Waal, F. B. M. (2001). Apes from Venus: Bonobos and human social evolution. In: F. B. M. de Waal (Ed.), Tree of origin: What primate behavior can tell us about human social evolution (pp. 41–68). Cambridge, MA: Harvard University Press. de Waal, F. B. M. (2006). Primates and philosophers: How morality evolved. Princeton, NJ: Princeton University Press. de Waal, F. B. M. (2010). The age of empathy: Nature’s lessons for a kinder society. New York: Harmony. de Waal, F. B. M., & Lanting, F. (1998). Bonobo: The forgotten ape. Berkeley, CA: University of California Press. de Waal, F. B. M., & van Hooff, J. A. R. A. M. (1981). Side-directed communication and agonistic interactions in chimpanzees. Behaviour, 77(3), 164–198. de Waal, F. B. M., & van Roosmalen, A. (1979). Reconciliation and consolation among chimpanzees. Behavioral Ecology and Sociobiology, 5, 55–66. de Waal, F. B. M., van Hooff, J. A., & Netto, W. J. (1976). An ethological analysis of types of agonistic interaction in a captive group of Java monkeys (Macaca fascicularis). Primates, 17(3), 257–290. Deets, A. C., & Harlow, H. F. (1971). Early experience and the maturation of aggression. In: V. P. Rock (Ed.), Value and knowledge requirements for peace. Washington, DC: AAAS. Diamond, J. (1992). The rise and fall of the third chimpanzee: Evolution and human life. London: Hutchinson. Dittus, W. P. J. (1977). Group fusion among wild Toque macaques: An extreme case of intergroup resource competition. Behaviour, 100(1–4), 247–291. Dolhinow, P., & Bishop, N. (1972). The development of motor skills and social relationships among primates through play. In: P. Dolhinow (Ed.), Primate patterns. New York: Holt, Rinehart & Winston.
86
JOHAN M. G. VAN DER DENNEN
Eberhart, J. A., & Candland, D. K. (1981). A preliminary model of primate intergroup encounters. In: P.F. Brain & D. Benton (Eds), The biology of aggression (pp. 577-584). Alphen a.d. Rijn: Sijthoff & Noordhoff. Ellefson, J. O. (1968). Territorial behavior in the common white-handed gibbon, Hylobates lar. In: P. C. Jay (Ed.), Primates: Studies in adaptation and variability (pp. 180–199). New York: Holt, Rinehart & Winston. Fashing, P. J. (2001). Male and female strategies during intergroup encounters in Guerezas (Colobus guereza): Evidence for resource defense mediated through males and a comparison with other primates. Behavioral Ecology and Sociobiology, 50(3), 219–230. Fossey, D. (1971). More years with mountain gorillas. National Geographic Magazine, 140(4), 574–585. Fossey, D. (1979). Development of the mountain gorilla (Gorilla gorilla berengei): The first thirty-six months. In: Hamburg & McCown (Eds), pp. 139–186. Fragaszy, D. M., Visalberghi, E., & Fedigan, L. M. (2004). The complete capuchin: The biology of the genus Cebus. Cambridge, UK: Cambridge University Press. Gabow, S. L. (1973). Dominance order reversal between two groups of free-ranging rhesus monkeys. Primates, 14(2–3), 215–223. Galdikas, B. M. F. (1979). Orangutan adaptation at Tanjung Puting reserve: Mating and ecology. In: Hamburg & McCown (Eds), pp. 195–234. Garber, P. A., Prietz, J. D., & Isaacson, J. (1993). Patterns of range use, range defense, and intergroup spacing in moustached tamarin monkeys (Saguinus mystax). Primates, 34(1), 11–25. Gat, A. (1999). The pattern of fighting in simple, small-scale, pre-state societies. Journal Anthropological Research, 55(4), 563–583. Givens, R. D. (1975). Aggression in nonhuman primates: Implications for understanding human behavior. In: M. A. Nettleship, R. D. Givens & A. Nettleship (Eds), War, its causes and correlates (pp. 263–280). The Hague: Mouton. € € € Tierpsychologie, 3, Goethe, F. (1939). Uber das Anstossnehmen bei Vogeln. Zeitschrift fur p. 371. Goodall, J. (1968). The behaviour of free-living chimpanzees in the Gombe Stream Reserve. Animal Behavior Monographs, 1(3), 161–311. Goodall, J. (1970). Tool using in primates and other vertebrates. Advanced Studies of Behavior, 3, 195–249. Goodall, J. (1971). Some aspects of aggressive behavior in a group of free-living chimpanzees. International Social Science Journal, 23(1), 89–97. Goodall, J. (1977). Infant killing and cannibalism in free-living chimpanzees. Folia Primatologica, 28, 259–282. Goodall, J. (1979). Life and death at Gombe. National Geographic Magazine, 155(5), 592–621. Goodall, J. (1986). The chimpanzees of Gombe: Patterns of behavior. Cambridge, MA: Harvard University Press. Goodall, J., et al. (1979). Intercommunity interactions in the chimpanzee population of the Gombe National Park. In: Hamburg & McCown (Eds), pp. 13–54. Hall, K. R. L. (1962). The sexual, agonistic and derived social behaviour patterns of the wild chacma baboon (Papio ursinus). Proceedings Zoological Society London, 139, 283–327. Hall, K. R. L. (1965). Behaviour and ecology of the wild patas monkey, Erythrocebus patas, in Uganda. Journal Zoological Society London, 14, 265–289.
The Biopolitics of Primates
87
Hall, K. R. L., & DeVore, I. (1965). Baboon social behavior. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 53–110). New York: Treubner King. Hamburg, D. A. (1972). Aggressive behavior of chimpanzees and baboons in natural habitats. Journal of Psychiatric Research, 8, 385–398. Hamburg, D. A. (1978). Human aggressiveness and conflict resolution. In: N. C. Dahl & J. B. Weisner (Eds), World change and world security (pp. 39–60). Cambridge, MA: MIT Press. Hamburg, D. A., & Goodall, J. (1974). Factors facilitating development of aggressive behavior in chimpanzees and humans. In: J. de Wit & W. W. Hartup (Eds), Determinants and origins of aggressU`ive behavior (pp. 59–85). The Hague: Mouton. Harcourt, A. H. (1978). Strategies of emigration and transfer by primates with particular € Tierpsychologie, 48, 401–420. reference to gorillas. Zeitschrift fur Harcourt, A. H., & de Waal, F. B. M. (Eds). (1992). Coalitions and alliances in humans and other animals. Oxford, UK: Oxford University Press. Hausfater, G. (1972). Intergroup behavior of free-ranging rhesus monkeys (Macaca mulatta). Folia Primatologica, 18, 78–107. Hausfater, G., & Hrdy, S. B. (Eds). (1984). Infanticide: Comparative and evolutionary perspectives. New York: Aldine. Higley, J. D. (2003). Aggression. In: D. Maestripieri (Ed.), Primate psychology (pp. 17–40). Cambridge MA: Harvard University Press. Hrdy, S. B. (1974). Male-male competition and infanticide among the langurs (Presbytis entellus) of Abu, Rajasthan, India. Folia Primatologica, 22(1), 19–58. Hrdy, S. B. (1977a). Infanticide as a primate reproductive strategy. American Scientist, 65(1), 40–49. Hrdy, S. B. (1977b). The langurs of Abu: Female and male strategies of reproduction. Cambridge, MA: Harvard University Press. Hrdy, S. B. (1979). Infanticide among animals: A review, classification, and examination of the implications for the reproductive strategies of females. Ethology and Sociobiology, 1, 13–40. Hrdy, S. B. (1981). The woman that never evolved. London: Harvard University Press. Huntingford, F. A., & Turner, A. (1987). Animal conflict. London: Chapman & Hall. Ihobe, H. (1992). Observations on the meat-eating behavior of wild bonobos (Pan paniscus) at Wamba, Republic of Zaı¨ re. Primates, 33, 247–250. Itani, J. (1982). Intraspecific killing among non-human primates. Journal of Social and Biological Structures, 5(4), 361–368. Itani, J., Tokuda, K., Furuya, Y., Kano, K., & Shin, Y. (1963). The social construction of natural troops of Japanese monkeys in Takasakiyama. Primates, 4(3), 1–42. Jay, P. C. (1965). The common langur in northern India. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 197–249). New York: Treubner King. Johnson, G. R. (1995). The evolutionary origins of government and politics. In: A. O. Somit & J. A. Losco (Eds), Research in biopolitics. Vol. 3: Human nature and politics (pp. 243–305). Greenwich CT: JAI Press. Jolly, A. (1966). Lemur behavior: A Madagascar field study. Chicago: University of Chicago Press. Jolly, A. (1972). The evolution of primate behavior. New York: Macmillan. Kano, T. (1984). Observations of physical abnormalities among the wild bonobos (Pan paniscus) of Wamba, Zaire. American Journal of Physical Anthropology, 63, 1–11.
88
JOHAN M. G. VAN DER DENNEN
Kano, T. (1987). Social regulation for individual coexistence in pygmy chimpanzees (Pan paniscus). In: D. McGuiness (Ed.), Dominance, aggression and war (pp. 105–118). New York: Paragon House. Kano, T., & Mulavwa, M. (1984). Feeding ecology of the pygmy chimpanzee (Pan paniscus) of Wamba. In: R. L. Susman (Ed.), The pygmy chimpanzee: Evolutionary biology and behavior (pp. 223–234). New York: Plenum Press. Kawai, M. (1964). The ecology of Japanese monkeys. Tokyo: Kawade-Shoko. Kawai, M. (1965). On the system of social ranks in a natural troop of Japanese monkeys on Koshima Islet. In: Imanishi & Altmann (Eds). Kawanaka, K. (1973). Intertroop relationships among Japanese monkeys. Primates, 14(2–3), 113–159. Kawanaka, K. (1981). Infanticide and cannibalism in chimpanzees, with special reference to the newly observed case in the Mahale Mountains. African Studies Monographs, 1, 69–99. King, G. E. (1980). Alternative uses of primates and carnivores in the reconstruction of early hominid behavior. Ethology and Sociobiology, 2, 99–109. Kinzey, W. G., & Robinson, J. G. (1983). Intergroup loud calls, range size, and spacing in Callicebus torquatus. American Journal of Physical Anthropology, 60, 539–544. Kitamura, K. (1983). Pygmy chimpanzee association patterns and ranging. Primates, 24, 1–12. Kitchen, D. M., & Beehner, J. C. (2007). Factors affecting individual participation in grouplevel aggression among non-human primates. Behaviour, 144, 1551–1581. Kitchen, D. M., Cheney, D. L., & Seyfarth, R. M. (2004). Factors mediating intergroup encounters in savannah baboons (Papio cynocephalus ursinus). Behaviour, 141, 197–218. Klein, L. L. (1974). Agonistic behavior in neotropical primates. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 77–122). New York: Academic Press. Knauft, B. M. (1991). Violence and sociality in human evolution. Current Anthropology, 32(4), 391–409. Kortlandt, A. (1965). How do chimpanzees use weapons when fighting leopards? Yearbook of the American Philosophical Society, 327–332. Kortlandt, A. (1975). Wild chimpanzees using clubs in fighting an animated stuffed leopard. In: M. A. Nettleship, R. D. Givens & A. Nettleship (Eds), War, its causes and correlates (pp. 297–298). The Hague: Mouton. Kortlandt, A. (1980). How might early hominids have defended themselves against large predators and food competition. Journal of Human Evolution, 9, 79–112. Kortlandt, A. (1999). Ape models of incipient hominid lifestyles: Chimpanzee or pygmy chimpanzee (bonobo)? In: H. Ullrich (Ed.), Hominid evolution: Lifestyles and survival strategies (pp. 25–43). Schwelm (Germany): Edition Archaea. Kruuk, H. (1972). The spotted hyaena: A study of predation and social behavior. Chicago: University of Chicago Press. Kruuk, H. (1975). Hyaena. Oxford, UK: Oxford University Press. Kummer, H. (1968). Social organization of Hamadryas baboons: A field study. Basel: Karger. € W., & Angst, W. (1974). Triadic differentiation: An inhibitory process Kummer, H., Gotz, protecting pair bonds in baboons. Behaviour, 49, 62–87. Kurland, J. A. (1977). Kin selection in the Japanese monkey. In: Contributions to primatology (Vol. 12). Basel: Karger. Kuroda, S. (1980). Social behavior of the pygmy chimpanzees. Primates, 21, 181–197.
The Biopolitics of Primates
89
Lasswell, H. D. (1936). Politics: Who gets what, when and how. New York: Free Press of Glencoe. Lazaro-Perea, C. (2001). Intergroup interactions in wild common marmosets, Callithrix jacchus: Territorial defense and assessment of neighbors. Animal Behavior, 62(Pt. 1), 11–21. Leighton, D. R. (1987). Gibbons: Territoriality and monogamy. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 135–145). Chicago: University of Chicago Press. Littlefield, B. L. (2010). Infanticide following male takeover event in Verreaux’s sifaka (Propithecus verreauxi verreauxi). Primates, 51(1), 83–86. Low, B. S. (1993). An evolutionary perspective on war. In: W. Zimmerman & H. K. Jacobson (Eds), Behavior, culture, and conflict in world politics (pp. 13–56). Ann Arbor, MI: University Michigan Press. MacKinnon, J. R. (1979). Reproductive behavior in wild orangutan populations. In: Hamburg & McCown (Eds), pp. 257–274. Manson, J. H., & Wrangham, R. W. (1991). Intergroup aggression in chimpanzees and humans. Current Anthropology, 32(4), 369–377. Mason, W. A. (1966). Social organization of the South American monkey Callicebus moloch: A preliminary report. Tulane Studies in Zoology, 13, 23–28. Mason, W. A. (1968). Use of space by Callicebus groups. In: P. C. Jay (Ed.), Primates: Studies in adaptation and variability (pp. 200–216). New York: Holt, Rinehart & Winston. Masters, R. D. (1975). Politics as a biological phenomenon. Social Science Information, 14, 7–63. Masters, R. D. (1983). The biological nature of the state. World Politics, 35(2), 161–193. Masters, R. D. (1989). The nature of politics. New Haven, CT: Yale University Press. McGrew, W. C. (1992). Chimpanzee material culture. Cambridge, MA: Cambridge University Press. McGrew, W. C. , Marchant, L. F., & Nishida, T. (Ed.) (1996). Great ape societies. Cambridge, UK: Cambridge University Press. McGrew, W. C., Tutin, C. E. G., Baldwin, P. J., Sharman, M. J., & Whiten, A. (1978). Primates preying upon vertebrates: New records from West Africa. Carnivore, 1, 41–45. McGrew, W. C., Tutin, C. E. G., & Baldwin, P. J. (1979). New data on meat eating by wild chimpanzees. Current Anthropology, 20(1), 238–239. McGuire, M. T. (1982). Social dominance relationships in male vervet monkeys: A possible model for the study of dominance relationships in human political systems. International Political Science Review, 3, 11–32. McKenna, J. J. (1982). Primate field studies: The evolution of behavior and its socioecology. In: J. L. Forbes & J. E. King (Eds), Primate behavior (pp. 53–83). New York: Academic Press. Mech, L. D. (1977). Productivity, mortality, and population trends of wolves in Northeastern Minnesota. Journal of Mammalogy, 58, 559–574. Miller, P. (1995). Crusading for chimps and humansyJane Goodall. National Geographic, 188(6), 102–128. Mitani, J. C., & Rodman, P. S. (1979). Territoriality: Relation of ranging patterns and home range size to defendability, with an analysis of territoriality among primate species. Behavioral Ecology and Sociobiology, 5, 241–251. Mitani, J. C., Watts, D. P., & Amsler, S. J. (2010). Lethal intergroup aggression leads to territorial expansion in wild chimpanzees. Current Biology, 20(12), R507–R508.
90
JOHAN M. G. VAN DER DENNEN
Mohnot, S. M. (1971). Some aspects of social changes and infant-killing in the hanuman langur Presbytis entellus (Primates: Cercopithecidae) in Western India. Mammalia, 35, 175–198. Morrison, J. A., & Menzel, E. W. (1972). Adaptation of a free-ranging rhesus monkey group to division and transportation. Wildlife Monographs, 31, 1–78. Muller, M. N., & Wrangham, R. W. (Eds). (2009). Sexual coercion in primates and humans: An evolutionary perspective on male aggression against females. Cambridge, MA: Harvard University Press. Nagel, U., & Kummer, H. (1974). Variation in cercopithecoid aggressive behavior. In: R.L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 159–185). New York: Academic Press. Nash, L. T. (1976). Troop fission in free-ranging baboons in the Gombe Stream National Park, Tanzania. American Journal of Physical Anthropology, 44, 63–78. Netto, W. J., & van Hooff, J. A. R. A. M. (1986). Conflict interference and the development of dominance relationships in immature Macaca fascicularis. J. G. Else & P. C. Lee (Eds), Primate ontogeny, cognition, and social behaviour (pp. 291–300). Cambridge, UK: Cambridge University Press. Nishida, T. (1979). The social structure of chimpanzees of the Mahale Mountains. In: Hamburg & McCown (Eds), pp. 73–121. Nishida, T. (1980). On inter-unit-group aggression and intra-group cannibalism among wild chimpanzees. Human Ethology Newsletter, 31, 21–24. Nishida, T. (1983). Alpha status and agonistic alliance in wild chimpanzees (Pan troglodytes schweinfurthii). Primates, 24(3), 318–336. Nishida, T. (1987). Local traditions and cultural transmission. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 462–474). Chicago: University of Chicago Press. Nishida, T., & Hiraiwa-Hasegawa, M. (1987). Chimpanzees and bonobos: Cooperative relationships among males. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 165–177). Chicago: University of Chicago Press. Nissen, H. W. (1931). A field study of the chimpanzee. Comparative Psychological Monographs, 8(1), 1–105. No€e, R., de Waal, F. B. M., & van Hooff, J. A. (1980). Types of dominance in a chimpanzee colony. Folia Primatologica, 34, 90–110. Nunn, C. L., & Deaner, R. O. (2004). Patterns of participation and free riding in territorial conflicts among ringtailed lemurs (Lemur catta). Behavioral Ecology and Sociobiology, 57, 50–61. Nunn, C. L., & Lewis, R. J. (2001). Cooperation and collective action in animal behaviour. € J. A. van Hooff & P. Hammerstein (Eds), Economics in nature: Social In: R. Noo; dilemmas, mate choice and biological markets (pp. 42–66). Cambridge: Cambridge University Press. Oates, J. F. (1977). The social life of the black-and-white colobus monkey, Colobus guereza. € Tierpsychologie, 45(1), 1–60. Zeitschrift fur Packer, C. R. (1977). Reciprocal altruism in Papio Anubis. Nature, 265, 441–443. Packer, C. R., & Pusey, A. E. (1982). Cooperation and competition within coalitions of male lions: Kin selection or game theory?. Nature, 296(5859), 740–742. Packer, C. R., & Pusey, A. E. (1983). Adaptations of female lions to infanticide by incoming males. American Naturalist, 121(5), 716–728.
The Biopolitics of Primates
91
Packer, C. R., & Pusey, A. E. (1984). Infanticide in carnivores. In: Hausfater & Hrdy (Eds), pp. 31–42. Packer, C. R., Scheel, D., & Pusey, A. E. (1990). Why lions form groups: Food is not enough. American Naturalist, 136, 1–19. Parish, A. R. (1994). Sex and food control in the ‘uncommon Chimpanzee’: How bonobo females overcome a phylogenetic legacy of male dominance. Ethology and Sociobiology, 15(3), 157–179. Parish, A. R. (1996). Female relationships in bonobos (Pan paniscus): Evidence for bonding, cooperation, and female dominance in a male-philopatric species. Human Nature, 7(1), 61–96. Parish, A. R., & de Waal, F. B. M. (1999). The other ‘closest living relative’: How bonobos (Pan paniscus) challenge traditional assumptions about females, dominance, intra- and intersexual interactions, and hominid evolution. In: D. LeCroy & P. Moller (Eds), Evolutionary perspectives on human reproductive behavior (pp. 97–113). New York: New York Academy of Sciences. Payne, H. F. P., Lawes, M. J., & Henzi, S. P. (2003). Competition and the exchange of grooming among female samango monkeys (Cercopithecus mitis erythrarchus). Behaviour, 140, 453–471. Peterson, S. A., & Somit, A. (1980). Cost-benefit analysis, shifting coalitions, and human evolution. Human Ethology Newsletter, 31, 16–18. Pitcairn, T. K. (1974). Aggression in natural groups of pongids. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 241–272). New York: Academic Press. Plavcan, J. M., van Schaik, C. P., & Kappeler, P. M. (1995). Competition, coalitions and canine size in primates. Journal of Human Evolution, 28, 245–276. Poirier, F. E. (1970). Nilgiri langur ecology and social behavior. In: L. A. Rosenblum (Ed.), Primate behavior: Developments in field and laboratory research (Vol. 1, pp. 103–135). New York: Academic Press. Poirier, F. E. (1974). Colobine aggression: A review. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 123–158). New York: Academic Press. Preston, S. D., & de Waal, F. B. M. (2003). Empathy: Its ultimate and proximate bases. Behavioral and Brain Sciences, 25(1), 1–34. Pride, R. E., Felantsoa, D., Randriamboavonjy, R., & Randriambelona, R. (2006). Resource defense in Lemur catta: The importance of group size. In: A. Jolly, R. W. Sussman, N. Koyama & H. Rasamimanana (Eds), Ringtailed lemur biology (pp. 208–232). Lemur catta, Madagascar. New York: Springer. Pruetz, J. D., & Bertolani, P. (2007). Savanna chimpanzees, Pan troglodytes verus, hunt with tools. Current Biology, 17(5), 412–417. Pusey, A. E. (1979). Intercommunity transfer of chimpanzees in Gombe National Park. In: Hamburg & McCown (Eds), pp. 465–480. Pusey, A. E. (2001). Of genes and apes: Chimpanzee social organization and reproduction. In: de Waal (Ed.), pp. 11–37. Pusey, A. E., Williams, J., & Goodall, J. (1997). The influence of dominance rank on the reproductive success of female chimpanzees. Science, 277(5327), 828–831. Rasmussen, D. R. (1981). Communities of baboon troops (Papio cynocephalus) in Mikumi National Park, Tanzania. Folia Primatologica, 36, 232–242.
92
JOHAN M. G. VAN DER DENNEN
Rawlins, R. G., & Kessler, M. (1986). The Cayo Santiago macaques. Albany, NY: SUNY. Reynolds, V., & Reynolds, F. (1965). Chimpanzees of the Budongo forest. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 368–424). New York: Treubner King. Ripley, S. (1967). Intertroop encounters among Ceylon grey langurs (Presbytis entellus). In: S. S. Altmann (Ed.), Social communication among primates (pp. 237–254). Chicago: University Chicago Press. Robinson, J. G. (1979). Vocal regulation of use of space by groups of titi monkeys Callicebus € Tiersychologie, 49, 381–405. moloch. Zeitschrift fur Robinson, J. G. (1981). Vocal regulation of inter- and intragroup spacing during boundary encounters in the titi monkey, Callicebus moloch. Primates, 22, 161–172. Robinson, J. G. (1988). Group size in wedge-capped capuchin monkeys Cebus olivaceus, and the reproductive success of males and females. Behavioral Ecology and Sociobiology, 23, 187–197. Rowell, T. E. (1972). The social behaviour of monkeys. Harmondsworth: Penguin. Rudran, R. (1973). Adult male replacement in one-male troops of purple-faced langurs (Presbytis senex senex) and its effect on population structure. Folia Primatologica, 19, 166–192. Saito, C., et al. (1998). Aggressive intergroup encounters in two populations of Japanese macaques (Macaca fuscata). Primates, 39(3), 302–312. Sanderson, I. T. (1955). Living mammals of the world. London: Hamish Hamilton. Schaller, G. B. (1963). The mountain gorilla: Ecology and behavior. Chicago: University of Chicago Press. Schaller, G. B. (1965). The behavior of the mountain gorilla. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 324–367). New York: Treubner King. Schenkel, R., & Schenkel-Hilliger, L. (1967). On the sociobiology of free-ranging colobus (Colobus guereza caudatus Thomas 1885). In: D. Starck , R. Schneider & H. Kuhn (Ed.), Progress in primatology. Stuttgart, Germany: Fischer. Schubert, G. (1983). Evolutionary politics. Western Political Quarterly, 36(2), 175–193. Schubert, G. (1986). Primate politics. Social Science Information, 25, 647–680. Schubert, G. (1991). Primate politics. In: G. Schubert & R. Masters (Eds), Primate politics (pp. 37–56). Carbondale, IL: Southern Illinois University Press. Schubert, G., & Somit, A. (Eds). (1982). The biology of primate sociopolitical behavior. DeKalb, IL: Northern Illinois University Press. Schuster, R. H. (1978). Ethological theories of aggression. In: I. L. Kutash , S. B. Kutash & L. B. Schlesinger (Ed.), Violence: Perspectives on murder and aggression (pp. 74–100). San Francisco: Jossey-Bass. Scott, J. P. (1969). Biological basis of human warfare: An interdisciplinary problem. In: M. Sherif & C. W. Sherif (Eds), Social psychology (pp. 121–136). New York: Harper & Row. Shea, B. T. (1983). Paedomorphosis and neoteny in the pygmy chimpanzee. Science, 22(2), 521–522. Silk, J. B. (1997). The function of peaceful post-conflict contacts among primates. Primates, 38, 265–279. Silk, J. B. (1998). Adaptive perspectives on conflict remediation in monkeys, apes, and humans. Human Nature, 9(4), 341–368. Silk, J. B. (2002a). The form and function of reconciliation in primates. Annual Review of Anthropology, 312, 21–44.
The Biopolitics of Primates
93
Silk, J. B. (2002b). Kin selection in primate groups. International Journal of Primatology, 23(4), 849–875. Silk, J. B., & Boyd, R. (1983). Cooperation, competition, and mate choice in matrilineal macaque groups. In: Wasser (Ed.), pp. 316–349. Slurink, P. (2002). Why some apes became humans: Competition, consciousness and culture. Nijmegen, The Netherlands: P.S. Press. Smuts, B. B., & Smuts, R. W. (1993). Male aggression and sexual coercion of females in nonhuman primates and other mammals: Evidence and theoretical implications. Advances in the Study of Behavior, 22, 1–63. Somit, A. (1990). Humans, chimps, and bonobos: The biological bases of aggression, war, and peacemaking. Journal of Conflict Resolution, 34(3), 553–582. Somit, A., & Peterson, S. A. (1995). Darwinism, dominance, and democracy. In: A. O. Somit & J. A. Losco (Eds), Research in biopolitics. Vol. 3: Human nature and politics (pp. 19–34). Greenwich CT: JAI Press. Somit, A., & Peterson, S. A. (1998). Biopolitics after three decades: A balance sheet. British Journal of Political Science, 28(3), 559–571. Southwick, C. H. (1962). Patterns of intergroup social behaviour in primates, with special reference to rhesus and howler monkeys. Annals, 102, 436–454. Southwick, C. H. (Ed.) (1963). Primate social behavior. Princeton, NJ: Van Nostrand. Southwick, C. H. (1967). An experimental study of intragroup agonistic behavior in rhesus monkeys. Behaviour, 28(1-2), 182–209. Southwick, C. A., Beg, M. A., & Siddiqi, M. R. (1965). Rhesus monkeys in North India. In: I. DeVore (Ed.), Primate behavior: Field studies of monkeys and apes (pp. 111–159). New York: Treubner King. Southwick, C. H., Siddiqi, M. R., Farooqui, M. Y., & Pal, B. C. (1974). Xenophobia among free-ranging rhesus groups in India. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 185–210). New York: Academic Press. Stanford, C. B. (1991). Social dynamics of intergroup encounters in the capped langur (Presbytis pileata). American Journal of Primatology, 25(1), 35–47. Stanford. (2001). The ape’s gift: Meat-eating, meat-sharing, and human evolution. In: de Waal (Ed.), pp. 97–118. Starin, E. D. (1994). Philopatry and affiliation among red colobus. Behaviour, 130(3–4), 253–270. Steenbeek, R. (1999). Tenure related changes in wild Thomas’s langurs. 1: Between-group interactions. Behaviour, 136(5), 595–625. Steenbeek, R., Piek, R. C., van Buul, M., & van Hooff, J. A. (2000). Vigilance in wild Thomas’s langurs (Presbytis thomasi): The importance of infanticide risk. Behavioral Ecology and Sociobiology, 45, 137–150. Stephenson, G. R. (1975). Social structure of mating activity in Japanese macaques. In: S. Kondo, M. Kawai, A. Ehara & S. Kawamura (Eds), Contemporary primatology. Basel, Switzerland: Karger. Strier, K. B. (2001). Beyond the apes: Reasons to consider the entire primate order. In: de Waal (Ed.), pp. 71–94. Strier, K. B., Ziegler, T. E., & Wittwer, D. J. (1999). Seasonal and social correlates of fecal testosterone and cortisol levels in wild male muriquis (Brachyteles arachnoides). Hormones and Behavior, 35(2), 125–134.
94
JOHAN M. G. VAN DER DENNEN
Struhsaker, T. T. (1969). Behavior of vervet monkeys and other cercopithecines. Science, 156, 1197–1203. Struhsaker, T. T. (1977). Infanticide and social organization in the redtail monkey € (Cercopithecus ascanius schmidti) in the Kibale Forest, Uganda. In: Zeitschrift fur Tierpsychologie, 45, 1, 1977, pp. 75–84, p. 591. Struhsaker, T. T., & Leland-Struhsaker, L. (1979). Socioecology of five sympatric monkey species in the Kibale Forest, Uganda. In: J. S. Rosenblatt, R. A. Hinde, E. Shaw & C. G. Beer (Eds), Advances in the Study of Behavior (Vol. 6, pp. 159–228). New York: Academic Press. Struhsaker, T. T., & Leland-Struhsaker, L. (1987). Colobines: Infanticide by adult males. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 83–97). Chicago: University of Chicago Press. Sugiyama, Y. (1967). Social organization in hanuman langurs. In: Altmann (Ed.), pp. 221–236. Sugiyama, Y. (1976). Life history of male Japanese monkeys. In: J. S. Rosenblatt, R. A. Hinde, E. Shaw & C. G. Beer (Eds), Advances in the Study of Behavior (Vol. 6, pp. 255–284). New York: Academic Press. Sugiyama, Y., Yoshiba, K., & Parthasarathy, M. D. (1965). Home range, mating season, male group, and intertroop relations in Hanuman langurs (Presbytis entellus). Primates, 6(1), 73–106. Surbeck, M., & Hohmann, G. (2008). Primate hunting by bonobos at Luikotale, Salonga National Park. Current Biology, 18, R906–R907. Surbeck, M., Fowler, A., Deimel, C., & Hohmann, G. (2009). Evidence for the consumption of arboreal, diurnal primates by bonobos (Pan paniscus). American Journal of Primatology, 71(2), 171–174. Sussman, R. W., & Richard, A. (1974). The role of aggression among diurnal prosimians. In: R. L. Holloway (Ed.), Primate aggression, territoriality, and xenophobia: A comparative perspective (pp. 49–76). New York: Academic Press. Suzuki, A. (1971). Carnivority and cannibalism observed among forest-living chimpanzees. Journal of the Anthropological Society Nippon, 79, 30–48. Tenaza, R. R. (1975). Territory and monogamy among Kloss’ gibbons (Hylobates klossii) in Siberut Island, Indonesia. Folia Primatologica, 24(1), 60–80. Thayer, B. A. (2004). Darwin and international relations: On the evolutionary origins of war and ethnic conflict. Lexington, KY: University Kentucky Press. Trudeau, M. B., Bergmann-Riss, E., & Hamburg, D. A. (1981). Towards an evolutionary perspective on aggressive behavior: The chimpanzee evidence. In: D. A. Hamburg & M. B. Trudeau (Eds), Biobehavioral aspects of aggression (pp. 27–40). New York: Alan R. Liss. Tutin, C. E. G. (1975). Sexual behaviour and mating patterns in a community of wild chimpanzees (Pan troglodytes schweinfurthii). Ph.D. Dissertation, University of Edinburgh. Van der Dennen, J. M. G. (Ed.) (1992). The nature of the sexes: The sociobiology of sex differences and the ‘battle of the sexes’. Groningen: Origin Press. Van der Dennen, J. M. G. (1995). The origin of war: The evolution of a male-coalitional reproductive strategy. Groningen: Origin Press. Van der Dennen, J. M. G. (2011). The biopolitics of great apes. In: A. Somit & S. A. Peterson (Eds), Biology and politics: Where it has been and where it is going. Opladen: Barbara Budrich.
The Biopolitics of Primates
95
Van Hooff, J. A. R. A. M. (1973a). The Arnhem Zoo chimpanzee consortium. International Zoo Yearbook, 13, 195–205. Van Hooff, J. A. R. A. M. (1973b). Sociale relaties bij dieren. In: G .P. Baerends et al. (Eds), Ethologie, de biologie van gedrag (pp. 223–256). Wageningen: Pudoc. Van Hooff, J. A. R. A. M. (1990a). Intergroup competition and conflict in animals and man. In: J. M. G. Van der Dennen & V. Falger (Eds), Sociobiology and conflict: Evolutionary perspectives on competition, cooperation, violence and warfare (pp. 23–54). London: Chapman & Hall. Van Hooff, J. A. R. A. M. (1990b). Groepsagressie en oorlog, een vergelijkend oecoethologische benadering. In: J. A. Van Hooff, G. Benthem van den Bergh & J. M. Rabbie (Eds), Oorlog; multidisciplinaire beschouwingen (pp. 21–75). Hoogezand: Stubeg. van Schaik, C. P. (2002). Primates: Primate societies and social life. In: M. Pagel (Ed.), Encyclopedia of evolution (pp. 934–939). Oxford: Oxford University Press. Van Schaik, C. P., & Janson, C. H. (Eds). (2000). Infanticide by males and its implications. Cambridge, UK: Cambridge University Press. Vervaecke, H. (2002). Bonobo’s: schalkse apen met menselijke trekjes. Leuven: Davidfonds. Vessey, S. H. (1968). Interactions between free-ranging groups of rhesus monkeys. Folia Primatologica, 8, 228–239. Vine, I. (1973). Social spacing in animals and man. Social Science Information, 12(5), 7–50. Walters, J. R., & Seyfarth, R. M. (1987). Conflict and cooperation. In: B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker (Eds), Primate societies (pp. 306–317). Chicago: University of Chicago Press. Waser, P. M., & Wiley, R. H. (1979). Mechanisms and evolution of spacing in animals. In: P. Marler & J. G. Vandenbergh (Eds), Handbook of behavioral neurobiology (Vol. 3). New York: Plenum Press. Washburn, S. L., & Hamburg, D. A. (1968). Aggressive behavior in Old World monkeys and apes. In: P. C. Jay (Ed.), Primates: Studies in adaptation and variability (pp. 458–478). New York: Holt, Rinehart & Winston. Wasser, S. K. (Ed.) (1983). Social behavior of female vertebrates. New York: Academic Press. Willhoite, F. H. (1976). Primates and political authority: A biobehavioral perspective. American Political Science Review, 70(December), 1110–1126. Williams, J. M., Oehlert, G., & Pusey, A. E. (2004). Why do male chimpanzees defend a group range? Reassessing male territoriality. Animal Behavior, 68, 523–532. Wilson, M. L. (2003). Environmental factors and aggression in non-human primates. In: M. P. Mattson (Ed.), Neurobiology of aggression: Understanding and preventing violence (pp. 151–165). Totowa, NJ: The Humane Press. Wilson, M. L., & Wrangham, R. W. (2003). Intergroup relations in chimpanzees. Annual Review of Anthropology, 32, 363–392. Wilson, M. L.Britton, N. F. & Franks, N. R. (2002). Chimpanzees and the mathematics of battle. Proceedings Royal Society London, B269, 1107–1112. Wilson, M. L., Wallauer, W. R. & Pusey, A. E. (2004). New cases of intergroup violence among chimpanzees in Gombe National Park, Tanzania. International Journal of Primatology, 25(3), 523–549. Wilson, W. L., & Wilson, C. C. (1968). Aggressive interactions of captive chimpanzees living in a semi-free-ranging environment. Technical Report, Aeromedical Research Lab, New Mexico.
96
JOHAN M. G. VAN DER DENNEN
Wrangham, R. W. (1979). On the evolution of ape social systems. Social Science Information, 18, 355–386. Wrangham, R. W. (1999). Evolution of coalitionary killing. Yearbook of Physical Anthropology, 42, 1–30. Wrangham, R. W. (2006). Why apes and humans kill. In: M. Jones & A. C. Fabian (Eds), Conflict: The 2005 Darwin college lecture series (pp. 43–62). Cambridge, UK: Cambridge University Press. Wrangham, R. W., & Peterson, D. (1996). Demonic males: Apes and the origins of human violence. Boston: Houghton Mifflin. Wrangham, R. W. (2001). Out of the ‘Pan’ into the fire: How our ancestors’ evolution depended on what they ate. In: F. B. M. de Waal (Ed.), Tree of origin: What primate behavior can tell us about human social evolution (pp. 121–143). Cambridge, MA: Harvard University Press. Wrangham, R. W., & Pilbeam, D. (2001). African apes as time machines. In: R. H. Tuttle, et al. (Eds), All apes great and small: African apes (Vol. 1, pp. 5–17). New York: Kluwer/ Plenum. Wright, P. C. (1978). Home range, activity pattern, and agonistic encounters of a group of night monkeys (Aotus trivirgatus) in Peru. Folia Primatologica, 29(1), 43–55. Yeager, C. P., & Kirkpatrick, R. C. (1998). Asian colobine social structure: Ecological and evolutionary constraints. Primates, 39(2), 147–155. Yerkes, R. M., & Yerkes, A. W. (1929). The great apes: A study of anthropoid life. New Haven, CT: Yale University Press. Zhao, Q. K. (1997). Intergroup interactions in Tibetan macaques at Mt. Emei, China. American Journal of Physical Anthropology, 104(4), 459–470.
MEASURING SOCIAL AND POLITICAL PHENOTYPES Levente Littvay INTRODUCTION Measurement is probably the most classical empirical process independent of time period or area of study. The title of this chapter suggests that a discussion of how to measure when approaching the borders of social and natural sciences is what follows, but this is somewhat misleading. Good measurement is independent of how the measures are used. Measures smothered with error, on the contrary, will be bad measures. A quick survey of the political science literature will paint a blatantly clear picture. The vast majority of empirical studies do not discuss or consider the implications of measurement decisions, namely, the reliability or validity of the measures. While I have a difficult time believing that researchers do not consider such issues, one could believe otherwise based on the presentations found in journals. Most researchers are aware of the difficulty of fitting an entire study into the 6,500 words requested by the journals. Decisions to cut certain areas are inevitable, and insufficient discussion of measurement decisions and their consequences are often one of the first casualties of the process. This is an unfortunate reality, and therefore, this chapter is devoted to the general assessment of measurement issues and measurement decision in empirical biopolitics.
Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 97–114 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009006
97
98
LEVENTE LITTVAY
The purpose of this chapter is to explore the consequences of what bad measures in research that borders the social and natural sciences, specifically within behavior genetics. It is important to think about what exactly we are interested in when we marry the social sciences with the natural sciences. The lack of such considerations will inevitably lead to poor measures and measurement decisions. We need to see through the lens of behavior genetics and develop constructs that get us closer to understanding social behavior in this framework. Finally, beyond a review of measurement issues within behavior genetics, the original contribution of the chapter describes the behavior genetic techniques that can aid our understanding of measurement quality from the perspective of reliability and validity. Are they good measures or are they not, and how can behavior genetic approaches help us answer this question?
REVIEW OF THE LITERATURE Behavior genetics has been a prevalent subfield in psychology (Jang, McCrae, Angleitner, Riemann, & Livesley, 1998; van Beijsterveldt & van Baal, 2002), but other social sciences such as political science have been slower to consider the possibility that genes can influence behaviors. Alford et al.’s American Political Science Review article exploring the genetic transmission of political attitudes laid the groundwork for this type of research within political science (Alford, Funk, & Hibbing, 2005). Findings of this article have appeared in popular media (such as The New York Times) and the article became the journal’s most downloaded piece. It was cited by the editor as the most important article published by the journal in its 100-year history (Hatemi, Mediland, Morley, Heath, & Martin, 2007). Since then, several pieces exploring how genetics could influence political behavior have appeared in top political science journals such as Political Analysis (Medland & Hatemi, 2009), the American Political Science Review (Fowler, Baker, & Dawes, 2008), the American Journal of Political Science (Hatemi et al., 2010), The Journal of Politics (Hatemi, Medland, & Eaves, 2009a; Hatemi, Funk, & Medlund, 2009b; Fowler & Dawes, 2008; Dawes & Fowler, 2009; Settle, Dawes, Christakis, & Fowler, 2010), Political Behavior (Sturgis et al., 2010), Canadian Journal of Political Science (Bell, Schermera, & Vernon, 2009). Political Research Quarterly published a special issue on the topic (Settle, Dawes, & Fowler, 2009; Hatemi, Hibbing, & Alford, 2009c), and Political Psychology is currently preparing its special issue. An article on the topic is also published in Science (Fowler & Schreiber, 2008).
Measuring Social and Political Phenotypes
99
Economists are also starting to publish work on genetic predictors of economic behavior in prestigious journals such as the Proceedings of the National Academy of Science (Cesarini et al., 2008) and The Quarterly Journal of Economics (Cesarini et al., 2009). The topic has also gained press within sociology (Guo & Stearns, 2002; Eaves et al., 1999; also see special issue on biological predictors of social behavior in Social Forces September 2006, Issue 85:1). Social Science Quarterly recently published a call to put together a special issue on the topic. Behavior geneticists have also developed an interest in the ‘‘phenotypes’’ studied by social scientists. The Behavior Genetics Association’s journal, Behavior Genetics, has already published articles on the topic of politics (Hatemi et al., 2007; Eaves & Hatemi, 2008) and also genes’ interactions with socioeconomic variables (Harden, Turkheimer, & Loehlin, 2007). Twin Research and Human Genetics, a leading journal of behavior genetics, is currently preparing its political phenotypes special issue. Behavior genetic exploration of social and political traits allowed us to better understand the nature of these political traits. But behavior genetic approaches can also help us understand social and political constructs from a different and heavily underappreciated perspective: the perspective of measurement reliability and validity.
RELIABILITY AND VALIDITY It is standard practice in psychology to present evidence that the constructs used in any study are both reliable and valid. These issues are discussed less within political science, though this does not diminish their importance. Reliability is defined as our ability to measure without error. Reliability assumes that what we want to measure has a true score (TS), and any unsystematic deviation from that TS in our measurement is random measurement error. The proportions of the TS variance and the error variance quantify reliability (Carmines & Zeller, 1979). Reliability can come in different forms. For example, interrater reliability tests if different people using the same measurement instrument (or guidelines) vary in their measurements. Test–retest reliability assesses the variances across two successive measurements temporally assuming no change in the TS is expected in between the two measurements. Turning to validity, a valid measure is one that measures what it is supposed to measure, to ‘‘meaningfully capture the ideas contained in the corresponding concept’’ (Adcock & Collier, 2001; see also Carmines & Zeller, 1979). Note the
100
LEVENTE LITTVAY
difference between reliability and validity. A reliable measure does not mean that a measure is valid and vice versa. It is possible to measure something with high reliability and completely misinterpret what it measures. Validity also comes in a large variety of forms, though Adcock and Collier argued that validity cannot be separated into various forms of validity, rather the forms should be seen as methods of validation that provide evidence for validity (Adcock & Collier, 2001). The forms of validation that closely resemble the proposed behavior genetic method are tests of construct and convergent validity. Construct validity builds on the intercorrelation between the indicators of a construct as a test of common underlying variance (Carmines & Zeller, 1979). Factor analytical approaches that test the validity of a factor underlying the indicators derive information from these intercorrelations. Convergent validity uses theoretically sound causes and consequences of the construct and correlates the measure of the construct with such causes and consequences (Carmines & Zeller, 1979). If the theoretically expected correlates actually correlate with the measure, that provides evidence for validity. Holding some notable exceptions, evidence is rarely presented that any measure used is reliable and valid. Interestingly, country level research within political science addresses questions of validity (see Adcock & Collier, 2001, for a discussion of validity in our democratic performance measures). This is especially odd as the social science that is most concerned with validity is psychology, a discipline driven by the exploration of individual differences. There are noteworthy exceptions in the political science literature that address issues of measurement, reliability and validity explicitly. For example, a reasonable amount of space is devoted to the discussion of measuring political knowledge (Mondak, 1999; Barabas, 2002; Sturgis, Allum, & Smith, 2008). Accuracy of measurement, issues of reliability and validity are not entirely ignored within political science; it is easier to find articles where these issues are not addressed at all, than to find ones where they are considered even in passing. Articles that substantively address these issues remain rare.
THE UNIVARIATE ACE MODEL To assess the genetic heritability most studies utilize a classical twin design. The model that derives the estimates of interest is a univariate structural equation model also presented in Fig. 1 (Neale & Cardon, 1992; Medland & Hatemi, 2009). Estimates of interest are the impact of the additive
101
Measuring Social and Political Phenotypes 1 1/.5
E1
C1
Phenotype Twin 1
A1
A2
C2
E2
Phenotype Twin 2
Fig. 1. The Univariate ACE Model.
genetic (A), common or shared environmental (C), and unique environmental (E) components generally, in its standardized form, expressed as proportions of variance explained in the trait. With certain assumptions, data from monozygotic (MZ) and dizygotic (DZ) twins raised together allow us to quantify these sources of variance. The result is A, C, and E estimates that, in standardized form, always sum to 1. Other possible sources of variance are assumed to be 0 in the classical twin model (though can be extended to test other quantities of interest as well). What allows us to derive such estimates is that MZ twins share their entire genome whereas DZ twins share, on average, 50% of their genome, like any two siblings. For this reason, if MZ twins are more similar to their co-twin for a particular trait than DZ twins, we can infer that there is a genetic basis for that trait. The structural equation model presented in Fig. 1 assumes that the distribution of the shared environment is the same for both MZ twins and DZ twins. This assumption is also known as the equal environment assumption (or EEA). The EEA received extensive criticism and sparked a debate on the theoretical level (Charney, 2008; Alford, Funk, & Hibbing, 2008a; Hannagan & Hatemi 2008; Beckwith & Morris, 2008, Alford, Funk, & Hibbing, 2008b), but a careful survey of the literature yielded no empirical study, to date, which demonstrated an EEA violation. Finally, the model allows for the existence of environmental influences that are not shared among the co-twins. In essence this category is the residual category known in the regression framework. In a twin model, deviations from perfect co-twin trait correlations for MZ twins suggest that the unique environment has an impact on the trait, but there is a caveat. Just like with regression residuals, another source of such deviations is measurement
102
LEVENTE LITTVAY
error. In fact, our understanding that measurement error is compounded in this residual category is what gives us the ability to make inferences about reliability, and to a lesser extent, validity. More formally speaking, the proportion of the trait variance explained by A, C, and E components are estimated with a two-group structural equation model presented in Fig. 1. A, C, and E are latent variables influencing the observed phenotype of one of the twins. They are latent, as we do not measure them directly, but make assumptions about their relationship to the same latent sources of the other twin of the pair. As seen in Fig. 1, A is perfectly correlated across MZ co-twins and 0.5 correlated for the DZ co-twins. This is the component that capitalizes on the difference between MZ and DZ twins and is the only difference between the two model groups (MZ and DZ). C is perfectly correlated for both MZ and DZ twins, as this is the environment that both twins share irrespective of their zygosity. E, the residual category, is unique to each and therefore is uncorrelated across the twins.
RELIABILITY TESTING WITH THE ACE MODEL There are multiple ways the univariate ACE model can contribute to our understanding of our measure’s reliability and validity. First, if reliability is depreciated with random measurement error and the unique environmental component collects all random measurement error, we are always safe to state that the maximum possible measurement error in a construct is less than the unique environmental component. For example, if a construct contains 20% random measurement error, at minimum the E component will be above 20%. How much above will be determined by the proportion of true unique environmental component. More formally, let us assume that a construct’s TS is driven by 40% additive genetic (tsA), 40% shared environmental (tsC), and 20% unique environmental (tsE) component. On top of the TS, our measure contains 20% measurement error (1TS). That means 80% of the construct is its TS. In this case, assuming accurate estimation, we would estimate: A ¼ TS tsA ¼ 0:8 0:4 ¼ 0:32 C ¼ TS tsC ¼ 0:8 0:4 ¼ 0:32 E ¼ TS tsE þ ð1 TSÞ ¼ 0:8 0:2 þ 0:2 ¼ 0:36 This means that we underestimate both additive genetic and shared environmental effect. We overestimate the unique environmental effect.
Measuring Social and Political Phenotypes
103
The mathematics presented clearly show that the estimated E cannot be smaller than the amount of measurement error. Even if the true unique environmental impact is 0 (tsE ¼ 0), the estimated E will be equal to the proportion of measurement error. On the basis of the above derivation, we can state with certainty that measurement error will never be larger than the estimated unique environmental component. From this univariate model we cannot infer exactly how much the measurement error is, but we do know that it cannot exceed the estimated unique environmental component. While this might not sound overly satisfying, if we estimate a very low unique environmental component, we can be confident that measurement error is not plaguing our estimate to a large extent. To demonstrate with an example, Hatemi et al. (2007) showed that the unique environmental component of party choice is 19%. It is reassuring that party choice is not measured with much, no more than 19%, measurement error. Of course, while we might think that we can reasonably expect people to correctly identify their party preference without too much error, a number of studies argued that party ID suffers from substantial measurement error (Clarke et al., 2004). On the contrary, if we are uncertain about the reliability of a measure (especially a single indicator measure), the availability of a twin sample allows us to make a statement, admittedly a vague one, about the measure’s validity. These statements about reliability are admittedly vague. Fortunately, the multivariate extension of the ACE model, as presented in the Cholesky Decomposition section below, allows us to partition the unique environmental component into more meaningful subcomponents. These subcomponents will allow us to make a more precise statement about measurement error. The precision of the statement will never be absolute and therefore only a plausible maximum amount of measurement error can ever be derived from such reliability analysis.
VALIDITY TESTING WITH THE ACE MODEL The univariate ACE model can provide us with information about the validity of our construct as well. Convergent validity tests whether certain causes and consequences of the construct in question correlate, as expected, with the constructs measures. Expected correlations can be used as evidence for validity. Conversely, if two measures of the same construct are driven by different sources of variation, that is evidence against validity. The ACE
104
LEVENTE LITTVAY
model’s job is to decompose the variation into various sources. If the ACE model shows that one measure of the construct is predominantly driven by additive genetic effects and the other measure is driven by shared environmental components, that is clear evidence against the validity of at least one of the measures and evidence that the measure is polluted by other systematic latent factors that do not meaningfully capture the construct we wish to capture. Unfortunately, the claim does not stand if the measure’s variation is driven by the same factors. Just because the shared environment predominantly drives both measures, we cannot claim that it is a strong evidence for validity. The shared environment is not a specific cause; rather, it is an additive collection of causes. For this reason it is still entirely possible that one of the measures is driven by a different source of the shared environment than the other measure. In the following section I discuss how a multivariate extension of the ACE model can refine the inferences presented here with the univariate model. First, let us discuss how the ACE model can be extended into a multivariate framework.
THE CHOLESKY DECOMPOSITION The most commonly used multivariate ACE model is the Cholesky decomposition (Neale & Cardon, 1992; Medland & Hatemi, 2009). In its bivariate form, the model estimates the additive genetic, shared and unique environmental effects of the first phenotype. It decomposes the covariation between the two phenotypes into covariation due to additive genetic, shared, and unique environmental covariation. Finally, it estimates the additive genetic, shared, and unique environmental effects of the second phenotype’s leftover variation not explained by the covariation with first phenotype. Owing to size limitations, Fig. 2 only shows the heritability model for twin 1. The same restrictions on the covariation between twins 1 and 2 presented in the univariate model are in place for A, C, and E. The Cholesky decomposition gives us additional leverage in separating the error variance from the TS variance of the trait. We already know from the univariate analysis that the measurement error of the trait is located in the E component, meaning technically the measurement error has to be less than E. The natural follow up question is how much less, and this is where the Cholesky decomposition helps us.
105
Measuring Social and Political Phenotypes
AI
CI
Phenotype I Fig. 2.
EI
A II
C II
E II
Phenotype II
Bivariate Cholesky Decomposition ACE Model for Twin 1.
Convergent validity assessment techniques validate measures through checking whether they are correlated with theoretical causes or consequences of the construct. If the correlations are observed as expected, that can be used as evidence for validity. The Cholesky decomposition also helps us understand the relationship between two variables (much like a correlation), but it goes beyond by providing us with the sources of the correlation. Correlation can come from common underlying additive genetic, common underlying shared environmental, and common underlying unique environmental effects. From the perspective of measurement error, we are predominantly interested in the common underlying unique environmental effect. Since measurement error is random, if all the E component in our measure is measurement error, it cannot, by definition, be correlated with anything. For this reason, true unique environmental component can be separated from measurement error if we observe common underlying unique environmental covariation with a trait that theoretically should correlate with our measure. For instance, if 30% of the total unique environmental component of our measure covaries with the unique environmental component of its predictor, we can assume that no more than 70% of the E component can be measurement error. It can also be informative to observe additive genetic and shared environmental covariation. Here, the most informative part of the model is the lack of additive genetic and shared environmental covariation. Assuming valid measures of the same constructs, common underlying additive genetic and shared environmental components of these two measures should largely overlap. If they do not overlap and either of the measures has unique variation explained by a sizable additive genetic or shared environmental component, that is evidence that the measure contains
106
LEVENTE LITTVAY
systematic variation unique to the measure and not part of the underlying construct. These components all provide evidence for a systematic latent factor polluting our measures that do not capture the construct of interest. These components all provide evidence against validity. So far, this chapter has presented specific ways that a behavior genetic model can inform us about the reliability and validity of our measures. In the remainder of the chapter, measurement is discussed in more general terms. The following section aims to highlight the importance of accurate measures specifically in the context of behavior genetic research and some methods that are found in the literature to improve our operationalization strategies and the quality of our measures.
IMPORTANCE OF MEASUREMENT Generally speaking, accurate measures are important for all the social sciences. Unfortunately, it is an issue that often does not receive adequate attention. When testing relationships between variables using correlations or regressions (as is often done in the social sciences), measurement error can allow for the emergence of biased results. With simple tests of differences or relationships, measurement error is not as problematic as it is with genetic models. Let us take two variables and call them IV (for independent variable) and DV (for dependent variable). If any of these two variables are measured with error, the correlation between IV and DV will be biased downward. This bias hinders our ability to find a significant relationship between IV and DV even if the relationship exists, therefore increasing the chance of committing a type II error. While we can normatively claim that all errors are bad, given the choice between type I and type II errors, most scientists would prefer to commit a type II error. Contrary to simple tests of relationships, the impact of measurement error in genetic models is not so straightforward. As the mathematical derivation explained above, measurement error produces different biases in our A, C, and E estimates. Much like with simple tests of differences or relationships, the estimate of interest will be biased downwards when assessing additive genetic (A) or common environmental (C) effects in a univariate ACE model. The true impact of the unique environment (E), on the contrary, will be biased upwards. An upward bias in the estimate increases the chances of committing a type I error. For more extended models (such as multivariate or extended family models), the impact of measurement error can be even
Measuring Social and Political Phenotypes
107
more complex and should be considered and discussed for each specific model used. The bias emerging from measurement error has been considered by a large number of studies. McCrae and Costa (2003) argued along these same lines concerning the heritability estimates of personality claiming that the well corroborated, roughly 50% heritability estimates might be underestimated. Medland and Hatemi (2009) make note of the biases that emerge from measurement error. Unfortunately, there are no easy solutions to overcome the problem of measurement error. The best solution is accurate measurement that requires solid understanding of our constructs and careful considerations of our measures. When measurement is developed with excessive attention to accuracy and compared to more haphazard measures of the same construct, the impact on heritability estimates is immediately visible. For example, Riemann et al. combined self-assessment of personality with peer ratings, arguably improving the quality of measurements (Riemann, Angleitner, & Strelau, 1997). Heritability estimates for Big 5 personality traits (openness, conscientiousness, extraversion, agreeableness, and neuroticism) have increased from 0.42–0.56 to 0.66–0.79 as the additional raters were considered. Unique environmental effects decreased accordingly as well. One workable strategy for more accurate measurement is the use of multiple indicators and latent variable models. If multiple indicators provide a valid measurement of an underlying latent variable, a confirmatory factor analysis (CFA) or item response theory (IRT) model can capture the common variation in the indicators and provide a phenotype that is free of the variation specific to the indicators (Long, 1983; de Ayala, 2009). Assuming valid indicators, this specific variation has to be driven by random measurement error. The ACE model has a commonly used multivariate extension that builds on these CFA models decomposing both the measurement error free latent factors and the variation specific to each indicator to additive genetic (A), shared (C), and unique (E) environmental variation. This model is called the common pathway model (Neale & Cardon, 1992; Medland & Hatemi, 2009). For dichotomous indicators, an IRT version of the ACE model is also becoming popular (van den Berg, Beem, & Boomsma, 2006; Fowler et al., 2008) Such approaches largely overcome the problem of the phenotypic measurement error that biases A and C estimates downwards and E estimates upwards. Solid understanding of the trait studied, multiple indicators, and models that attempt to eliminate measurement error before the phenotypic variance decomposition are key to accurate estimation within the twin modeling
108
LEVENTE LITTVAY
paradigm. Accuracy of estimates is ensured at the design phase of these studies and not during the analysis. If we do not have reliable and valid measures, multiple indicators, and theoretical causes and consequences of the trait of interest measured also in a reliable and valid way, we have little opportunity to derive accurate estimates from these models. It is crucial that these issues are considered at the design phase of such studies.
LIMITATIONS OF THE TWIN MODELS Behavior genetic heritability assessment is not as clear-cut as described above. The above paragraphs were presented as if we can make an accurate point estimate inference for the population of the unique environment, or other components of the ACE model. We obviously cannot as our estimates are prone to sampling error. Every inference comes with uncertainty statistics usually expressed in the forms of confidence intervals. These measures of uncertainty need to be considered when interpreting inferences in the matter described in this chapter. For example, if the unique environmental component of phenotype X is 30% (with 95% confidence interval of 25–35%), we cannot say that measurement error accounts for no more than 30% of the trait. We can only say that we are 95% confident that the trait contains no more than 35% measurement error. Fortunately, since most inference on reliability and validity of traits relies on the unique environmental estimate, the confidence intervals for these estimates tend to be relatively narrow. This is due to the nature of the ACE model that estimates unique environmental variation with high confidence, less the case for the additive genetic or shared environmental components. Of course, this does not mean that no bias is present in the unique environmental estimate. In fact, all measurement error biases this component, and therefore, unique environmental variation cannot be interpreted as true unique environmental variation, but the sum of all random measurement error and the impact of the unique environmental factors. Although the classical twin design is highly powerful in estimating the unique environmental component, more power is always desirable. Additional power can be derived from larger samples or more extended family designs. Power analyses show that the inclusion of different-sex twins and additional family members can substantially increase power (Posthuma & Boomsma, 2000). Other benefits include the ability to test genetic and environmental sources of sex differences and the relaxation of certain assumptions of the classical twin model. The most common extended design
Measuring Social and Political Phenotypes
109
includes different twins and non-twin siblings. More extended designs can also include both parents and offspring of the twins (Hatemi et al., 2010). The extended family design is also useful in that it further decomposes the sources of variation. The classical twin design only gives us estimates for additive genetic, common and unique environmental components. The extended family design decomposes these effects further by clarifying the mechanism of cross-generational transmission of the trait. What this means in terms of convergent validity is variation attributable to more variance sources, leading to better inferences from the perspective of validity. One of the great limitations of the validation with the twin design is that, while variance attributed to vastly different sources is clear evidence against validity, variance attributed to one source, such a common underlying shared environmental variance, is not sufficient evidence for validity. While the covariation is attributed to the same source, the ‘‘shared environment’’ is sufficiently broad to be tautological. It would be incorrect to claim that measure X is environmental and measure Y is environmental; therefore, the same thing causes them both. That is obviously conceptual stretching. Evidence toward validity becomes stronger the more we decompose the variation and covariation into narrower and narrower common sources. After a discussion of the limitations of the classical twin model as a method of inferring reliability and validity, and possible ways to overcome these limitations, the following section discusses the cutting edge of behavior genetics to find the most meaningful constructs (phenotypes) to consider in our models of social and political behavior.
ENDOPHENOTYPES Psychiatric genetics is going through a crisis as it has become painfully clear that we are never going to find that one gene for autism, schizophrenia, etc. Geneticists had to get comfortable with the idea that most psychiatric and behavioral traits are hugely polygenic, a realization that has sparked an entire reconsideration of disease and behavior in general. Maybe the reason we are not finding what we are looking for is because we not defining, operationalizing, and measuring the right trait that underlies the behavior of interest. This process of reconsideration within psychiatry has spawned the concept of endophenotypes (Gottesman & Gould, 2003). Endophenotypes can be thought of as the decomposition of behavioral symptoms into more stable causes with clearer physiological and genetic foundations. It can also be thought of as steps backwards in the physiological
110
LEVENTE LITTVAY
pathway that begins with the protein coding of gene expression that, through the complex neurological pathway, manifests itself as behavior. It is still unclear if the search for the endophenotypes of heritable behavioral traits will fulfill its promises of clarifying the mechanism from genes to behavior. Can it identify specific genes that more directly contribute to the variance of the endophenotype as opposed to the actual behavioral if interest. While I am personally not optimistic, the cause is noble and the trend cannot be ignored by political science. Even if endophenotypes do not bring us closer to understanding the physiology of politial behavior, they can help us think about our constructs and measurements of our constructs. One of the most frequently asked questions of people first exposed to the behavior genetics of political behavior, especially voting behavior is the following: how can voting behavior evolve when we have only been voting for a couple of generations? The question is fair and the answer deserves careful consideration. While people have not been voting for more than a century or two, certain human conditions that could influence a person’s decision to vote have existed since the times of hunter gatherer societies. What exactly are these human conditions? In finding the answer, it does not help that we are barely able to explain a third of variation in voting behavior with the classical behavioral kitchen sink of IV approach (Plutzer, 2002). At the same time, evidence from multiple replications shows that turnout is highly heritable (Fowler et al., 2008). These two pieces of evidence and the theoretical question presented above drove political scientists to look for predictors of voting behavior in places where they had previously not been looking. If turnout is heritable and classical social scientific predictors do not explain much of turnout, maybe we just need to look for causes elsewhere, perhaps among highly heritable psychological traits. This is the progression that lead to the extensive consideration of personality as a predictor of turnout (Mondak, Hibbing, Canache, Seligson, & Anderson, 2010). Sadly, personality is not delivering the kind of predictive power we would expect based on the high heritability estimates, instead it looks like another behavioral predictor that only explains a percent or two of the variation. Personality can hardly be considered an endophenotype. Psychiatric endophenotypes are rather ‘‘neurophysiological, biochemical, endocrinological, neuroanatomical, cognitive, or neuropsychological,’’ and personality does not fit any of these criteria (Gottesman & Gould, 2003). But, it is a start down the complex pathway, mechanism that explains our behavior. Unfortunately, we are doing very poorly when we try to explain political behavior through these physiological mechanisms. Political science is decades behind psychology and psychiatry, but hopefully many important
Measuring Social and Political Phenotypes
111
contributions are ahead of us in these areas as we try to explain certain political behaviors. Until then, we can and we should turn to more traditional predictors of political behavior that are heavily studied by behavioral genetics. Personalities seem like an obvious first choice, especially in relation to highly heritable political phenotypes. The search must not stop here, however. It is entirely possible that IQ has an impact on political knowledge, that general sense of control influences political efficacy (Blais & St-Vincent, 2010), that sense of duty drives turnout, altruism impacts contributions (Frost Keller, Hatemi, & Eaves, 2010) etc. Until we know neurophysiological or even cognitive mechanisms that take place when people are voting, volunteering, making political decisions, etc., we will not know what endophenotypes to look for, what paths to consider and what to test from the perspective of behavior genetics.
CONCLUSION The chapter was devoted to the issue of measurement when studying political phenotypes in the behavior genetic framework. It predominantly focused on studies using twin designs with possible additional family members added but pointed toward the best designs for specific gene hunting methodologies. Measurement is a key component of all empirical research. Within political science insufficient attention is devoted to the subject. Reliability and different forms of validation are at our disposal to better understand how good our measures are. This study’s original contribution to the field is the discussion on how twin methodology can help us understand our measures better. Behavior genetic studies express sensitivity to the fact that the twin design can aid our understanding of reliability and validity. Indeed, they often even demonstrate this fact (Riemann et al., 1997). On the contrary, to the best of the author’s knowledge, no study, to date, proposed the twin design explicitly as a method of reliability testing and validation. Next, the study elaborated on the importance of good measures in general, but especially within the behavioral genetic framework. In behavior genetics, measurement error not only causes underestimated parameters and type II error, for some estimates the bias is in the opposite direction and the probability of type I error becomes a direct positive function of the amount of measurement error. Finally, the study discussed the phenomenon of endophenotypes as a method to improve our operationalization of behavioral constructs of
112
LEVENTE LITTVAY
interest. While endophenotypic studies fell short of meeting the promise they offered for researchers’ inability to identify certain genes influencing various behaviors, the logic of these studies could be utilized within political science. To date, our field does not come close to understanding the physiological foundations of various behaviors as well as psychology, for example. Psychology’s constructs and methods can bring us baby steps closer to a more complete understanding of political behavior.
ACKNOWLEDGMENTS I thank the editors for asking me for this contribution, my mentors John Hibbing and Bob Belli. Colleagues, Peter Hatemi, Chris Dawes, Zoli Fazekas, Sebi Popa, and Paul Weith for the countless invaluable conversations about the theoretical issues presented in this chapter. Finally, thanks to my wife, Megan Thornton, for the never-ending support and readily available language editing.
REFERENCES Adcock, R., & Collier, D. (2001). Measurement validity: A shared standard for qualitative and quantitative research. American Political Science Review, 95(3), 529–546. Alford, J., Funk, C., & Hibbing, J. (2008a). Beyond liberals and conservatives to political genotypes and phenotypes. Perspectives on Politics, 6(2), 321–328. Alford, J., Funk, C., & Hibbing, J. (2008b). Twin studies, molecular genetics, politics, and tolerance: A response to Beckwith and Morris. Perspectives on Politics, 6(4), 793–797. Alford, J. R., Funk, C. L., & Hibbing, J. R. (2005). Are political orientations genetically transmitted? American Political Science Review, 99(2), 153–167. Barabas, J. (2002). Another look at the measurement of political knowledge. Political Analysis, 10(2), 209–233. Beckwith, J., & Morris, C. A. (2008). Twin studies of political behavior: Untenable assumptions? Perspectives on Politics, 6(4), 785–791. Bell, E., Schermera, J. A., & Vernon, P. A. (2009). The origins of political attitudes and behaviours: An analysis using twins. Canadian Journal of Political Science, 42, 855–879. Blais, A., & St-Vincent, S. L. (2010). Personality traits, political attitudes and the propensity to vote. European Journal of Political Research, doi:10.1111/j.1475-6765.2010.01935.x. Available at http://onlinelibrary.wiley.com/doi/10.1111/j.1475-6765.2010.01935.x/pdf Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment. Sage University Paper Series on Quantitative Applications in the Social Sciences, Sage Newbury Park, CA. Cesarini, D., Dawes, C. T., Fowler, J. H., Johannesson, M., Lichtenstein, P., & Wallace, B. (2008). Heritability of cooperative behavior in the trust game. Proceedings of the National Academy of Sciences, 105(10), 3721–3726.
Measuring Social and Political Phenotypes
113
Cesarini, D., et al. (2009). Genetic variation in preferences for giving and risk taking. Quarterly Journal of Economics, 124(2), 809–842. Charney, E. (2008). Genes and ideologies. Perspectives on Politics, 6(2), 299–320. Clarke, H. D., et al. (2004). Political choice in Britain. New York: Oxford University Press. Dawes, C. T., & Fowler, J. (2009). Partisanship, voting, and the dopamine D2 receptor gene. The Journal of Politics, 71(3), 1157–1171. de Ayala, R. J. (2009). The theory and practice of item response theory. New York: The Guilford Press. Eaves, L. J., Heath, A., Martin, N., Maes, H., Neale, M., Kendler, K., Kirk, K., & Corey, L. (1999). Comparing the biological and cultural inheritance of personality and social attitudes in the Virginia 30000 study of twins and their relatives. Twin Research, 2(1), 62–80. Eaves, L. J., & Hatemi, P. K. (2008). Transmission of attitudes toward abortion and gay rights: Parental socialization or parental mate selection? Behavior Genetics, 38(3), 247–256. Fowler, J., Baker, L., & Dawes, C. T. (2008). Genetic variation in political participation. American Political Science Review, 102(2), 233–248. Fowler, J. H., & Dawes, C. T. (2008). Two genes predict voter turnout. The Journal of Politics, 70(3), 579–594. Fowler, J. H., & Schreiber, D. M. (2008). Biology, politics, and the emerging science of human nature. Science, 322, 912–914. Frost Keller, A., Hatemi, P. K., & Eaves, L. (2010). The source of donations to political and charitable groups. Presented at the International Society of Political Psychology, San Francisco, CA. Gottesman, I. L., & Gould, T. D. (2003). The endophenotype concept in psychiatry: Etymology and strategic intentions. The American Journal of Psychiatry, 160, 636–645. Guo, G., & Stearns, E. (2002). Social influence on the realization of genetic potential for intellectual development. Social Forces, 80(3), 881–910. Hannagan, R. J., & Hatemi, P. K. (2008). The threat of genes: A comment on Evan Charney’s ‘genes and ideologies’. Perspectives on Politics, 6(2), 329–337. Harden, K. P., Turkheimer, E., & Loehlin, J. C. (2007). Genotype by environment interaction in adolescents’ cognitive aptitude. Behavior Genetics, 37(2), 273–283. Hatemi, P. K., Mediland, S. E., Morley, K. I., Heath, A. C., & Martin, A. G. (2007). The genetics of voting: An Australian twin study. Behavior Genetics, 37(3), 435–448. Hatemi, P. K., Medland, S. E., & Eaves, L. J. (2009a). The genetic source of variance in the gender gap. The Journal of Politics, 71(1), 262–276. Hatemi, P. K., Funk, C. L., & Medlund, S,E. (2009b). Genetic and environmental transmission of political attitudes over a life time. Journal of Politics, 71(4), 1141–1156. Hatemi, P. K., Hibbing, J., & Alford, J. (2009c). Is there a ‘‘party’’ in your genes? Political Research Quarterly, 62(3), 584–600. Hatemi, P. K., et al. (2010). Not by twins alone: Using the extended twin family designed to investigate the genetic basis of political beliefs. American Journal of Political Science, 54(3), 798–814. Jang, K. L., McCrae, R. R., Angleitner, A., Riemann, R., & Livesley, W. J. (1998). Heritability of facet-level traits in a cross-cultural twin sample: Support for a hierarchical model of personality. Journal of Personality and Social Psychology, 74(6), 1556–1565. Long, J. S. (1983). Confirmatory factor analysis: A preface to LISREL. Newbury Park, CA: SAGE.
114
LEVENTE LITTVAY
McCrae, R. R., & Costa, P. T., Jr. (2003). Personality in adulthood: A five-factor theory perspective (2nd ed.). New York: The Guilford Press. Medland, S. E., & Hatemi, P. K. (2009). Political science, biometric theory, and twin studies: A methodological introduction. Political Analysis, 17(2), 191–214. Mondak, J. J. (1999). Reconsidering the measurement of political knowledge. Political Analysis, 8(1), 57–82. Mondak, J. J., Hibbing, M. V., Canache, D., Seligson, M. A., & Anderson, M. R. (2010). Personality and civic engagement: An integrative framework for the study of trait effects on political behavior. American Political Science Review, 104(1), 85–110. Neale, M., & Cardon, L. (1992). Methodology for genetic studies of twins and families. Boston, MA: Kluwer Academic Publishers. Plutzer, E. (2002). Becoming a habitual voter: Inertia, resources, and growth in young adulthood. American Political Science Review, 96(1), 41–56. Posthuma, D., & Boomsma, D. I. (2000). A note on the statistical power in extended twin designs. Behavior Genetics, 30(2), 147–158. Riemann, R., Angleitner, A., & Strelau, J. (1997). Genetic and environmental influences on personality: A study of twins reared together using the self- and peer report NEO-FFI scales. Journal of Personality, 65(3), 449–475. Settle, J. E., Dawes, C. T., & Fowler, J. H. (2009). The heritability of partisan attachment. Political Research Quarterly, 62(3), 601–613. Settle, J. E., Dawes, C. T., Christakis, N. A., & Fowler, J. H. (2010). Friendships moderate an association between a dopamine gene variant and political ideology. Journal of Politics, 72(4), 1189–1198. Sturgis, P., Allum, N., & Smith, P. (2008). An experiment on the measurement of political knowledge in surveys. Public Opinion Quarterly, 72(1), 90–102. Sturgis, P., Read, S., Hatemi, P. K., Zhu, G., Trull, T., Wright, M. J., & Martin N. G. (2010). A genetic basis for social trust?. Political Behavior, 32(2), 205–230. van den Berg, S. M., Beem, L., & Boomsma, D. I. (2006). Fitting genetic models using Markov chain Monte Carlo algorithms with bugs. Twin Research and Human Genetics, 9(3), 334–342. Van Beijsterveldt, C. E., & van Baal, G. C. (2002). Twin and family studies of the human electroencephalogram: a review and a meta-analysis. Biological Psychology, 61, 111–138.
POLITICAL SCIENCE AND BEHAVIOR GENETICS: RETHINKING FOUNDATIONAL ASSUMPTIONS Evan Charney INTRODUCTION Political scientists have taken up behavior genetics (BG) at a momentous time in the science of genetics. Momentous, because the science of genetics is undergoing a paradigm shift [Petronis, A. (2010). Epigenetics as a unifying principle in the aetiology of complex traits and diseases. Nature, 465(7299), 721–727]. This shifting paradigm poses a significant challenge to both the prevailing methodologies of behavior genetics – twin, family, adoption studies – and one of the most noteworthy findings to emerge from such studies, that is, which we can call the principle of minimal parental effects. This is the supposition that the effect of the shared parental rearing environment on the behavioral phenotypes of offspring is statistically equivalent to zero (Plomin & Daniels, 1987). It is not uncommon nowadays to find twin, adoption, and family studies utilized in the study of political behavior (e.g., Alford, J., Funk, C. L., & Hibbing, J. R. (2005). Are political orientations genetically transmitted? American Political Science Review, 99(2), 153–167.); likewise, the principle of minimal parental effects is frequently invoked in such studies (e.g., Mondak, J. J., Hibbing, M. V., Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 115–138 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009007
115
116
EVAN CHARNEY
Canache, D., Seligson, M. A., & Anderson, M. A. (2010). Personality and civic engagement: An integrative framework for the study of trait effects on political behavior. American Political Science Review, 104(1), 85–110.). As we shall see, the challenge comes from recent discoveries in genetics that are radically transforming our understanding of the genome and its relationship to environment.
THE CHANGING CONCEPTION OF THE GENOME Mitochondrial DNA While we are accustomed to talk of the human genome, humans in fact possess at least two distinct genomes: nuclear DNA (nDNA), the familiar double stranded helix located in the cell’s nucleus, typically equated with the genome simpliciter, and mitochondrial DNA (mtDNA) – a circular structure of double stranded DNA consisting of 37 genes located within the mitochondria, membrane-enclosed intracellular organelles. Mitochondria, in addition to having the most essential function of supplying cells with energy, play a critical role in a host of metabolic processes and organ systems including, in particular, the nervous system. Mitochondrial DNA is inherited in a maternal manner, that is, via the oocytes, which contain anywhere from 11,000 to over 900,000 copies of mtDNA, while sperm contain only about 100 mtDNAs that are destroyed early in embryogenesis. The existence of numerous multiple copies of mtDNA in a single cell, known as polyploidy, is found not only in oocytes, but in all cells in the body, and varies by cell type, for example, 1,075–2,794 copies of mtDNA per cell in muscle cells, 1,200–10,800 in neurons, and up to 25,000 in liver cells (Clay Montier, Deng, & Bai, 2009). In addition to polyploidy, mtDNA exhibits heteroplasmy, the occurrence of allelic differences between the multiple copies of mtDNA in the same individual. There is now considerable evidence that mtDNA sequence heteroplasmy exists at an appreciable frequency in healthy subjects, and that it segregates differentially in different tissues (He et al., 2010). Furthermore, neither the amount nor location of heteroplasmy is ‘‘fixed.’’ Since the mtDNA mutation rate is anywhere from B9 to 25 W nDNA, mtDNA mutations can be inherited both somatically and via the germline (Lynch, Koskella, & Schaak, 2006). The association between human mtDNA mutations and phenotype is significantly complicated by both polyploidy and heteroplasmy. Due to heteroplasmy, not all of the copies of mutated
117
Political Science and Behavior Genetics
mtDNA will exist in all of the copies of the mtDNA genome. Whether or not a phenotype associated with mtDNA is expressed depends upon both whether the mutated mtDNA is localized in certain cells and tissues of the body and if so, which cells and tissues; and also on whether the amount of mutated mtDNA surpasses a certain ‘‘threshold’’ level necessary for the appearance of the phenotype (Taylor & Turnbull, 2005). What determines the amount of mutated mtDNA that a given offspring inherits via the germline to begin with? The answer is, the individual oocyte the offspring happened to develop from. The implications of the transmission of mtDNA for heritability estimates are as follows: The amount of mtDNA any given sibling possesses is a result both of whatever maternal oocyte she happened to develop from, as well as a host of other stochastic processes that occur during embryogenesis. Hence, siblings will not possess 50% of their mother’s mtDNA. Furthermore, during embryogenesis, maternal mtDNA is stochastically partitioned during mitotic cellular division post fertilization. Hence, although MZ twins (MZT) develop from a single fertilized oocyte, their mtDNA will after the first mitotic cellular division. Although MZTs do not have identical mtDNA, nonetheless, they develop from a single oocyte, whereas DZTs develop from two separate oocytes. What are the implications of this? Consider that any two oocytes may contain anywhere from 11,000 to W900,000 copies of mtDNA. It is possible, therefore, that MZTs on average share greater amounts of mutated mtDNA, but they will not have identical mtDNA, nor does there exist any fixed ratio for the amount of mtDNA MZTs are likely to share relative to DZTs. Hence, mtDNA can be a cause of greater phenotypic concordance between MZTs relative to DZTs, as well as a cause of phenotypic discordance between MZTs. On the other hand, mtDNA can be a cause of significant discordance among DZTs and siblings. Inasmuch as they develop from two separate fertilized oocytes, one DZT or nontwin sibling may develop from an oocyte containing 11,000 copies of mtDNA while the other sibling develops from an oocyte containing 900,000 copies of mtDNA, that is, one sibling may inherit a small amount of mutated mtDNA, while the other sibling inherits a large amount. .
Copy-Number Variations Until recently, single nucleotide polymorphisms (SNPs) were thought to be the predominant form of genomic variation and to account for most normal, as well as abnormal, phenotypic variation in humans (Gibbs, 2003).
118
EVAN CHARNEY
A recent and important development in human genetics has been the discovery of the ubiquitousness of a variety of structural variations in DNA – deletions, insertions, and duplications, as well as more complex multi-site variants of varying sizes – that change the chromosomal architecture. From 12 to 30% of the human genome is comprised of CNVs, and over 41% of all CNVs identified thus far overlap with known genes, suggesting that CNVs play a substantial role – at least as great as SNPs – in modulating gene expression (Fanciulli, Petretto, & Aitman, 2010). CNVs can alter the expression of genes located in the region of variable copy number, allowing expression levels to be higher or lower than could be achieved by a single copy number per haploid gene. CNVs can be inherited via the germline in the manner of polymorphisms, but variation also arises at a significant rate in somatic cells. Estimates of the mutation rates of CNVs range from 100 to 10,000 WSNPs across the human genome. Although persons have different forms and amounts of mtDNA in different cells and tissues of their bodies, it has been a long-standing dogma in molecular genetics that humans possess identical nDNA in all somatic cells (minus certain immune cells) (Dear, 2009). Somatic mosaicism is defined as the presence of populations of somatic cells with genetically distinct nuclear genomes in a single organism. Somatic mosaicism results from postzygotic mutations that are usually propagated to only a subset of adult cells. Mosaicism can occur, however, in germline cells as well, the latter being more accurately characterized as ‘‘germline mosaicism,’’ although somatic and germline mosaicism are not mutually exclusive and can be difficult to distinguish in practice. Germline mosaicism, as distinguished from somatic mosaicism, has the potential of being transmitted to offspring (Rodrı´ guez-Santiago et al., 2010). Until recently, somatic mosaicism was associated with pathogenesis, but there is growing evidence that it is common, if not the norm. Rodrı´ guezSantiago et al. (2010), as part of a genome-wide survey of mosaic genomic variation analyzed blood or buccal DNA samples of 1,991 adult individuals, and found copy-number mosaicism in 1.7% of tissues samples from the same individual, including 11 large (1.5–37 Mb) CNVs. Liang, Conte, Skarnes, and Bradley (2008) demonstrated that CNVs involving gains or losses of millions of base pairs occurs during mitotic divisions of mouse embryonic stem (ES) cells during routine culture involving relatively few cellular divisions. Luo et al. (2009), employing single-cell analysis, reported high copy-number variation between individual sperm cells obtained from healthy male donors, thereby indicating mosaicism in the germline. The apparent high cellular frequency of CNVs detected in these studies and their
Political Science and Behavior Genetics
119
presence in normal adult individuals suggests that this type of mosaicism is a widespread phenomenon in the human genome, and should be considered in the expanding repertoire of inter- and intraindividual genetic variation (Mkrtchyan et al., 2010). If somatic mosaicism occurs in healthy individuals during embryogenesis, then it should occur in healthy MZTs as well. Bruder (2008) confirmed this in a study of 29 pairs of MZTs using peripheral-blood derived DNA. They discovered large-scale CNV variations among all the twins, healthy and diseased, with an estimated frequency of up to 10% variation per twin pair. It should be emphasized that these variations were found in a single tissue – blood – so that the actual total percentage is likely much higher, and it is for this reason that Bruder noted (2008, p. 766): ‘‘It is likely that the confirmed CNVs shown here represent the ‘tip of the iceberg’ of all CNVs that are actually present in the studied twins.’’ It is important to keep in mind that CNVs are the only kind of structural variation common in DNA. For example, there are W1 million microsatellites, variable numbers of 1–6 bp repeats totaling o200 bp in length, that account for 3% of the DNA sequence; B150,000 minisatellites and variable number tandem repeats, polymorphic sequences containing 20–50 copies of 6–100 bp repeats; and B1 million deletions and insertions (InDel) of DNA segments. There is little doubt that what we know about structural genomic variation within and between individuals itself represents the ‘‘tip of the iceberg.’’
Transposable Elements One striking finding to emerge from the completion of the human genome sequencing project is that an astonishing 55% of the genome is composed of transposable elements (TEs), repetitive mobile DNA sequences – ‘‘jumping genes’’ – dispersed throughout the genome (Gibbs, 2003). There are many different kinds of TEs. One variety, know as retrotransposons, comprise 45% of the genome, and move by a replicative, ‘‘copy and paste’’ process: They copy themselves to RNA, while the original DNA copy is maintained at the same location (Goodier & Kazazian, 2008). The RNA copy is then reverse-transcribed into DNA and the DNA is inserted into the genome at a new location. Thus, these elements expand in number as they retrotranspose, leading to an increase in genomic DNA content. Conventional wisdom holds that TEs are ancient ‘‘selfish’’ DNA ‘‘parasites,’’ living within the genomes of cellular organisms in a manner
120
EVAN CHARNEY
analogous to viruses (Dawkins, 2006). And indeed, the insertion of TEs can disrupt gene expressivity and cause a variety of pathogenic conditions. Over time, mutations and truncations of TEs have rendered many of them inactive, and many retrotransposons have been rendered incapable of retrotransposition, that is, they are ‘‘transpositionally incompetent,’’ and those that remain competent are often associated with pathogenesis and normally silenced by epigenetic ‘‘gene-silencing’’ mechanisms (Callinan & Batzer, 2006). However, this one-sided view of retrotransposons as selfish DNA parasites is giving way to a much more complex picture. Long interspersed nucleotide elements (L1s) are a major group of retrotransposons that retain the capacity to retrotranspose, consisting of approximately 600,000 copies and occupying B18% of the mammalian genome. Current estimates are that the human genome contains B150 retrotranspositioncompetent (RC) L1s (RC-L1s), and approximately 10% of these elements are classified as highly active or ‘‘hot’’ (Brouha, 2003). A growing body of evidence indicates that retrotransposons constitute major structural variants in the human genome, can act as alternate gene promoter regions, appear to play a critical role in normal human development (and neural development in particular), and that individual variation in retrotransposition potential makes an important contribution to human genetic diversity (Huang et al., 2010). That a retrotransposon is transpositionally incompetent does not mean that it is incapable of affecting gene expression. Transpositionally incompetent L1s have been shown to act as alternate promoter regions for more than 40 human protein coding genes (Matlik, Redik, & Speek, 2006). A genome-wide scan has identified 23,000 candidate regulatory regions derived from retrotransposons, suggesting that retrotransposition expression has a key influence on the expressional output of the mammalian genome (Faulkner et al., 2009). L1s also enable the retrotransposition of other mobile DNA elements that require the functional products of L1s in order to move about the genome, such as Alu elements. There are W1 million Alu elements in the human genome and they account for about 11% of its mass. The Alu family is comprised of primate-specific short interspersed elements (SINEs), and constitutes the most prevalent repetitive element in the human genome. Recent research indicates that during embryogenesis, Alu elements play a critical role in the positioning of nuclesomes, the basic unit of DNA packaging, consisting of a segment of DNA wound around a histone protein core. The activity of L1s capable of retrotransposition is regulated by the epigenome (Ostertag & Kazazian, 2001), which regulates the expressional activity of genes by a number of processes, one of which is DNA
Political Science and Behavior Genetics
121
methylation.1 During gamete formation and in early embryogenesis, short waves of demethylation in a region of the L1 that serves as a promoter of expression allows L1s to escape epigenetic silencing. Preimplantation mouse embryos in which the expression of L1 retrotransposons is blocked exhibit developmental arrest (Beraldi, Pittoggi, Sciamanna, Mattei, & Spadafora, 2006). Coufal et al. (2009) demonstrated that human neural progenitor cells (NPCs), whether derived from fetal brains or from cultured ES cells, support the retrotransposition of an introduced human L1. NPCs into which the L1s were introduced generated distinct types of neurons (out of the estimated W10,000 morphologically distinct types of neurons in the human brain). This neural retrotransposition appeared to occur frequently and independently in individual neural precursor cells, resulting in a substantial number of newly transposed L1 elements in differentiated neurons: B80–800 new L1 insertions per cell were observed in some brain regions. These data show that L1s introduced into human NPCs can retrotranspose in the genome, giving rise to functional neurons. Although retrotransposition occurs in many different cell types, it appears to play an especially significant role in neuronal development. Using advanced DNA analysis, Coufal et al. (2009) detected more L1s in the genome of adult human brain cells than in the genomes of heart or liver cells from the same individual – as many as 100 extra copies per neuron tested relative to other organs – which is consistent with neural precursor cells having undergone more retrotransposition events than other tissues. Since L1 is a retrotransposon, each cycle of retrotransposition adds a new copy of L1 DNA at different positions in neuronal DNA, resulting in neuronal somatic mosaicism. Such neuron-to-neuron variation in genomic DNA content has been observed in both rodents and humans (Coufal et al., 2009), and a number of studies have shown that L1 insertions into neurons alter neuronal DNA expression levels. Given that changing the firing patterns of single neurons can have marked effects on behavior, and that single-gene alterations can cause profound effects in subpopulations of neurons, it is likely that some L1 insertions will have significant effects on the structure and function of the human brain (Martin, 2009). Most human L1 retrotransposition events appear to occur during embryonic development, and only a fraction of these events get into the germline and are heritable. However, studies indicate that L1 RNA – the RNA to which restrotransposons copy themselves in the process of retrotransposition – is heritable via the germline. Although it has been widely accepted that there is little, if any, RNA in mature sperm, a study in mice has demonstrated strong evidence of non-Mendelian inheritance of
122
EVAN CHARNEY
RNA molecules via the gametes of both genders. Kano et al. (2009) reported that the RNA of L1 elements is abundant in both germ cells and embryos, and showed that L1 RNA transcribed in male or female germ cells can be carried over through fertilization and integrate in daughter cells during embryogenesis, creating a de novo L1 retrotransposition event in the genome of offspring resulting in somatic mosaicism. Hence, meiotically stable and heritable RNA can cause a phenotypic change in offspring in the absence of the inducing parental allele. New neurons are continually generated throughout adulthood, predominantly in the dentate gyrus, a region in the hippocampus, and the subventricular zone, a brain structure situated in the wall of the lateral ventricle. These newly generated neurons form synapses and are functionally integrated into existing hippocampal neuronal circuits, and there is evidence that adult neurogenesis is important for synaptic plasticity, learning, and memory (Neves, Cooke, & Bliss, 2008). The level of adult hippocampal neurogenesis is positively and negatively modulated by environmental conditions, including neuronal activity, stress, and aging (Ma et al., 2009). Because neurogenesis occurs in certain regions of the brain throughout life, we would expect that adult NPCs would be able to support L1 retrotransposition, and this has recently been demonstrated in rodents (Muotri, Zhao, Marchetto, & Gage, 2009). Exercise is known to have a significant impact upon hippocampal neurogenesis: It significantly increases the amount of brain-derived neurotropic factor (BDNF) in the hippocampus, a protein that supports the survival of existing neurons and encourages the growth and differentiation of new neurons and synapses. Environmental events known to increase hippocampal neurogenesis would also be expected to increase L1 retrotransposition. Muotri et al. (2009) demonstrated that the effects of voluntary running in rodents doubled L1 somatic retrotransposition in the rodent brain. However, L1 retrotransposition was increased not only in neurogenic areas, but also in nonneurogenic areas, such as the cerebellum. This finding indicates that running not only increases the number of new L1 insertions in the brain, but also activates silenced L1 insertions in other nonneurogenic brain regions. To the extent that the environment influences levels of L1 retrotransposition, this influence could be mediated by epigenetic or hormonal mechanisms (or both). Depending upon its impact upon the brain, L1-induced somatic variability could either increase the risk of neurological disease, or induce behavioral changes that could help the organism to adapt better to changing environments (Marchetto, Gage, & Muotri, 2010).
Political Science and Behavior Genetics
123
Because where retrotransposons are activated in neural precursor cells ultimately end up in the brain is largely a result of stochastic factors (or perhaps, ‘‘controlled stochasticity’’), as is the amount of L1 RNA inherited from any given germ cell, the brains of MZTs will exhibit retrotransposon induced intertwin (in addition to intratwin) genetic heterogeneity (Marchetto et al., 2010). Such heterogeneity will result not only from brain-wide retrotransposition during embryogenesis, but from ongoing neurogenesis in the hippocampus that occurs throughout life. It is entirely conceivable that differences in lifestyle, such as one twin exercising more than another, could further contribute to an already existing neuronal genetic heterogeneity.
Aneuploidy Humans possess 23 pairs of chromosomes (diploidy), and any deviation from diploidy, involving either more or less chromosomes, is called aneuploidy. In humans, aneuploidy is typically associated with pathogenesis. For example, trisomies are disorders characterized by an extra chromosome and are generally classified by the chromosomal location, for example, Trisomy 21 (Down’s syndrome) and Trisomy 18 (Edward’s syndrome), while monosomies are characterized by the absence of a chromosome, for example, Turner’s syndrome is characterized by a single X chromosome instead of an XX or XY. In addition to conditions in which there exists a uniform variation in chromosomal number in all cells, aneuploidy may be present in some cells and not others, and/or the kind of aneuploidy can differ in different cells, giving rise to chromosomal somatic mosaicism (CSM). CSM is found in many of the same pathogenic conditions characterized by aneuploidy, as well as other pathogenic conditions (Modi, Berde, & Bhartiya, 2003). Despite this association with pathogenesis, there is a surprisingly high degree of CSM in the normal embryonic and adult human. Recent conservative estimates place the overall percentage of aneuploid neural cells in the adult brain at an astonishing 10%, involving monosomy, trisomy, tretrasomy, polyploidy, and uniparental disomy (Yurov et al., 2007). Given an estimated 100 billion neurons, this yields a rough (conservative) estimate of 10 billion neurons and 100–500 billion glial cells with one or another form of chromosomal aneuploidy. It is estimated that roughly 28% of embryonic neural precursor cells exhibit chromosomal aneuploidy in one form or another (Iourov et al., 2009). This chromosomal
124
EVAN CHARNEY
diversity appears to result from a high frequency of stochastic postzygotic chromosome mutations in somatic cells. Neuroprogenitor cells in numerous regions of the embryonic brain display cell division ‘‘defects’’ in normal cells that result in aneuploid adult cell progeny. Various lines of evidence indicate that brain tissues may be more prone to aneuploidy than other tissues (Rehen, 2005). Mature aneuploid neurons are functionally active and integrated into brain circuitry, showing distant axonal connections (Kingsbury, 2005). What are the consequences of functionally active aneuploid neurons that are integrated into brain circuitry? One likely outcome is neuronal signaling differences caused by altered gene expression, as documented in mammalian neural cells. Thus, a network composed of intermixed diploid and aneuploid neurons might produce unique signaling properties distinct from a network composed purely of diploid cells. It has also been proposed that aneuploidy in the normal brain serves as a possible mechanism for neuronal cell diversification, in a manner similar to that proposed for retrotransposons (Muotri et al., 2009). Once again, the highly stochastic nature of chromosomal aneuploidy and resulting CSM insures that MZTs will have differences in their neuronal DNA, and that siblings will depart from a presumed 50% possession of the parental nuclear genome.
THE EPIGENOME For the most part, genes are transcribed to produce RNA and proteins, but before a gene can be transcribed, it must be ‘‘turned on,’’ that is, activated. Genes are not ‘‘self-activating,’’ and the mere presence of a gene as part of an individual’s genotype does not entail that it is capable of being transcribed. Rather, genes are turned on and off by the epigenome, the complex biochemical regulatory system that silences, activates, and changes the expressional activity of genes without any change to the DNA sequence itself (Jirtle & Skinner, 2007). Because epigenetic changes can be inherited, the epigenome has been characterized as a nonDNA form of inheritance (Holliday, 1994). The three most extensively studied epigenetic mechanisms to date are DNA methylation, histone modification, and micro-RNAs. DNA methylation involves the addition of a methyl compound (CH3) onto cytosine bases, one of the four bases that make up the DNA sequence, commonly at sites where a cytosine base is followed by a guanine base (a ‘‘CpG’’ site), reducing the accessibility of DNA to expression factors.
Political Science and Behavior Genetics
125
Generally, high levels of methylation are associated with the silencing of gene expression, while active genes generally have lower levels of methylation. Histone modification encompasses several structural modifications of histones, specialized proteins around which DNA is wrapped to form chromatin, the complex combination of DNA, RNA, and protein that make up the chromosome. Gene expression can occur only when the DNA is unwrapped from the histone proteins and its nucleic acid sequences are exposed. Histone modification alters the structure of chromatin in such a way as to either increase or decrease the ability of expression factors to access regulatory sites on the DNA. Micro-RNAs (miRNA) are a class of noncoding RNAs (ncRNAs), RNA molecules that, unlike the more familiar messenger RNAs (mRNA) involved in the process of transcribing DNA, do not code for gene expression. Instead, they act as epigenetic regulatory elements that control gene expressivity by blocking mRNA from being translated, and are associated with gene silencing. The epigenetic regulatory system is essential to almost every aspect of human life, from embryogenesis to aging. Development itself is largely an epigenetic process, inasmuch as the epigenome plays a critical role in the differentiation of ES cells into distinct cells and tissues of the body. Epigenetic regulation or ‘‘programming’’ of DNA in the cells and tissues of the body can be stable throughout the lifetime of an organism, as in the normal programming involved in cellular differentiation, but it can also change in response to cellular signals that are driven by environmental events, a process that continues throughout the organism’s lifetime. Such environmentally driven epigenetic changes entail a corresponding epigenetic reprograming of gene expressivity (Sweatt, 2009). These changes can be lifelong and have significant consequences for a wide range of behavioral phenotypes (Champagne & Curley, 2009). It is this capacity to respond dynamically to environmental signals, inducing long-term changes in gene expressivity, this is one of the noteworthy features of the epigenome. Epigenetic changes can be inherited both via the germline and somatically (Jablonka & Raz, 2009). Somatic or mitotic epigenetic inheritance occurs when somatic cells inherit the epigenetic marking of their predecessors, and transmit these, through mitosis, to subsequent generations of cells. While these epigenetic changes bring about phenotypic changes in the organism, to the extent that they are limited to somatic cells and do not affect the germ line, they cannot be transmitted to progeny. Meiotic or transgenerational epigenetic inheritance, involves epigenetic changes to the DNA in the organism’s germ cells; these epigenetic changes, and the corresponding phenotype, are inherited by offspring and can be passed on from generation to
126
EVAN CHARNEY
generation in a non-Mendelian manner and without any changes to the DNA sequence. MZTs have been shown to exhibit significant epigenetic differences, and there is growing evidence that these epigenetic differences play a role in phenotypic differences, as well as phenotypic concordances, in MZTs (Kaminsky et al., 2009). Epigenetic discordance in MZTs is known to be associated with two developmental processes that are in large part epigenetically regulated: X-chromosome inactivation and genomic imprinting. Whether an allele is silenced from the maternally or paternally derived X chromosome is primarily a stochastic process, with maternally and paternally derived X chromosomes inactivated at approximately the same frequency. However, about 15% of X-linked genes escape inactivation altogether allowing for greater gene expressivity in these regions, and an additional 10% show variable patterns of inactivation and are expressed to different degrees (Carrel & Willard, 2005). Another 5–20% of healthy females with no family history of X-linked disorders exhibit ‘‘skewed’’ X-chromosome inactivation, in which either the maternal or the paternal chromosome is preferentially silenced. Patterns of X-inactivation and variable X-linked gene expressivity differ among female MZTs. Studies have indicated that skewed X-inactivation patterns are more frequent in MZTs than in DZTs or singletons (Trejo et al., 1994), and that monochorionic MZTs have much more similar X-chromosome inactivation patterns than dichorionic MZTs. Fraga et al. (2005) found that 19% of healthy female MZTs had skewed X-chromosome inactivation patterns that differed from that of their cotwin, and numerous studies have identified skewed X-inactivation in female MZ twin pairs discordant for X-linked conditions such as fragile X-syndrome (Kruyer et al., 1994). Differences in genomic imprinting may also be a cause of phenotypic discordance among MZ cotwins. Beckwith–Wiedemann syndrome (BWS) is an imprinting disorder associated with abnormalities of chromosome 11p15, and among MZTs discordant for BWS, one twin shows the imprinting defect on this gene whereas the unaffected twin does not (Weksberg et al., 2003). In addition to epigenetic variations associated with X-inactivation and imprinting, a number of studies have found significant differences in DNA methylation and histone modification profiles in whole genome-wide scans, select tissues samples, and specific genomic regions in MZ twin pairs (Kaminsky et al., 2009). Significant differences have been identified in MZ twin pairs as young as five (Wong et al., 2010); longitudinal studies have shown that these differences change over time. Comparisons between MZTs and DZTs cotwins matched for age and sex have also revealed significantly
Political Science and Behavior Genetics
127
higher differences in DNA methylation in the buccal cells of DZTs as compared to MZTs (Kaminsky et al., 2009). Higher discordances in DNA methylation among DZTs could be the result of greater differences in their DNA, at least in part, for while there are numerous examples of changes in methylation status without changes in DNA sequence, it is still possible that DNA allelic variation could be associated with certain specific epigenetic profiles. But there is evidence for the greater concordance in methylation status in MZTs vs. DZTs not being tied to greater MZ genetic similarity. Kaminsky et al. (2009) measured the levels of DNA methylation at 2,176 unique sites in the brain of two stains of inbred mice as compared with two strains of outbred mice, and concluded that the impact of DNA polymorphisms on methylation status did not appear to be significant. Greater discordances in DZT methylation profiles may be due to their having originated from two distinct oocytes and two distinct sperm bearing different epigenetic markings (as opposed to MZTs, which are presumed to originate from a single ovum and sperm). Studies have shown that in healthy human subjects, the majority of sperm cells of the same individual exhibit unique DNA methylation profiles (Flanagan et al., 2006). Male germ cells undergo unique and extensive epigenetic remodeling soon after their specification in ES cells and during the differentiation process to become mature spermatozoa (Seki et al., 2005). There are several lines of evidence that the epigenetic markings in sperm can influence embryonic development. For example, analysis of DNA methylation in sperm has identified hypomethylated promoters that reveal patterns of methylation similar to those found in ES cells (Farthing et al., 2008). These findings also suggest that epigenetic markings may be the cause of both phenotypic concordances and discordances.
IMPLICATIONS To assess the implications of the kinds of genetic and epigenetic heterogeneity and liability to change, we have examined for behavioral genetics, and in particular for the manner in which heritability estimates are derived, consider the following three assumptions: Assumption 1. Siblings share on average 50% of their parents’ (nuclear) DNA, half siblings 25% of their parental DNA, etc. Assumption 2. MZTs have identical (nuclear) genomes, that is, they share 100% of their nDNA, while DZTs share 50% of their nDNA.
128
EVAN CHARNEY
Assumption 3. The genetic identity – the structure of the DNA sequence – of persons never changes (with the exception of cancerous mutations). It remains ‘‘fixed’’ from the moment of conception (rates of gene expression, however, may change). ‘‘Since genes are fixed, they represent the purest measure of biological inheritance, virtually unaffected by environment and able to be collected throughout a person’s life’’ (Fowler & Dawes, 2008). Twin and adoption studies depend, foundationally, upon the truth of these three assumptions. Assumptions 1 and 2 regarding percentage of genomic relatedness provide the necessary ratios that form the biometric core of twin and adoption studies. Assumption 3 ensures that these ratios do not change over the lifetime of individuals. In light of the developments we have considered in molecular genetics, principles 1–3 are invalid: Siblings cannot be assumed to share 50% of their parents (nuclear) DNA; MZTs cannot be assumed to share 100% of their nDNA while DZTs share 50%; and the genome is subject to a good deal of change, both during embryogenesis and throughout the life course of the individual. Given this, heritability estimates derived from twin and adoption studies are in doubt. I would like to consider a possible objection/responses to this claim: The processes of genetic variation I have considered are somatic, that is, they are zygotic or de novo occurrences, rather than germline or meiotic, and hence, are not inherited parentally. In fact, this is not the case, inasmuch as these same processes have been shown to occur in germ cells as well. However, let me put this point aside for the moment. The objection would be that inheritance, and presumably heritability estimates, should refer to what is inherited via the germline, and the genome at the moment of conception (when maternal and paternal chromosomes fuse) is what is inherited in this manner. If what we are interested in is the amount of phenotypic variance that can be attributed to genotypic variance, then it is not clear on what basis we can privilege the genotype at the moment of conception, as opposed to at the eight-cell stage of the embryo, or the early blasotcyst, or at whatever point in development and throughout the life-course genotypic variation results in phenotypic variation. Drawing a line between the genotype at the moment of conception and every moment thereafter is to arbitrarily ‘‘fix’’ the genome at a given point in time, and to ignore the fact that what the genome becomes in the process of development is as important as what it was at the moment of conception. The impulse to draw such a distinction, however, may derive more from a view about the nature of the genome (or ‘‘the genome as it
Political Science and Behavior Genetics
129
exists at the moment of conception’’). This is the idea that the ‘‘genome’’ in some way determines, or ‘‘sets the range of possible outcomes,’’ for whatever genotypic transformations it (the genome) can or will undergo. And what underlies this view is both a conception of the genome as possessed of almost supernatural powers, and a crudely deterministic conception of the progression from genotype to phenotype. To be sure, the genotype at the moment of conception ‘‘determines,’’ for example, that a polymorphism of gene G exists at a certain locus on chromosome N. But it neither ‘‘determines’’ nor ‘‘sets the range of possible outcomes’’ for the many stochastic processes it will undergo throughout the developmental process. It neither determines, nor sets the range of possible outcomes, for the outcome that as a result of chromosomal aneuploidy there will one copy of chromosome N in neurons of type A, three copies in neurons of type B, and four copies in neurons of type C; that de novo somatic copynumber variations on the promoter region of G in glial cells in the amygdala will alter its expression rate from G genes without this copy-number variation; that retrotransposition in neural precursor cells will result in an increase in G’s expression rates in neurons of type H in the hippocampus, or that epigenetic effects due to factors x, y, and z in the embryonic environment will effectively silence G in neurons of type K in the cerebellum. These are postconception, zygotic transformations of the genome. What of the transformations that occur in the germs cells themselves prior to conception? It will be recalled that one of the distinguishing features of mtDNA is the presence of significant interindividual variation in the amount of mtDNA contained in any given oocyte as a result of oogenesis (anywhere from 11,000 to over 900,000 copies of mtDNA). We also saw that sperm from the same individual display significant epigenetic, copy-number, and L1 RNA variations, that is, germline mosaicism. What are the implications of such mosaicism? It entails at least the following: We can assume neither that every sperm contains 50% of paternal DNA (i.e., the static genome as it is presumed to exist in identical forms in all of the father’s cells), nor that every oocyte contains 50% of maternal DNA (ditto). And this entails that we cannot assume that offspring will contain 50% of their parental DNA.
BEHAVIORAL EPIGENETICS In rodents, maternal care is characterized by a complex series of behaviors. Of particular significance for offspring development are licking and grooming (LG) behaviors. Mothers exhibit stable interindividual differences
130
EVAN CHARNEY
in the degree of LG behavior they exhibit during lactation (the first week postpartum). Those whose frequency of LG is one SDWthe mean for any given cohort are conventionally designated ‘‘high LG’’; those scoring one SDothe mean are designated ‘‘low LG’’ (Szyf, Weaver, Champagne, Diorio, & Meaney, 2005). Whether a pup is nurtured by a high- or low-LG mother is associated, as an adult, with a range of endocrine, neurological, and behavioral outcomes. In general, high LG is associated with a range of behaviors we would be inclined to classify as ‘‘positive’’ behavioral outcomes, and low LG with ‘‘negative’’ outcomes. In mammals, a central component of the response to stress involves the hypothalamic–pituitary–adrenal (HPA) axis: The hypothalamus releases corticotrophin releasing factor (CRF), which stimulates the release of adrenocorticotropic hormone (ACTH) from the pituitary glands into the general bloodstream. ACTH in turn stimulates the release of the glucocorticoid hormones (cortisol in humans, corticosterone in rodents), and the neurotransmitters/hormones epinephrine and norepinephrine (the catecholamines). Pups raised by high-LG dams consistently exhibit, as adults, lower levels of stress, as measured both physiologically and behaviorally, relative to low-LG offspring. High-LG pups display reduced plasma levels of ACTH and corticosterone, and higher levels of gammaaminobutyric acidA (GABAA) neural receptors, which serve to inhibit corticotrophin releasing factor activity. Behaviorally, they exhibit decreased startle response, lower fearfulness, and increased exploration. Pups raised by low-LG dams exhibit, as adults, the opposite of all of these phenotypes. Furthermore, the female offspring of high-LG mothers become, as adults, high-LG mothers themselves, while the female offspring of low-LG mothers become low-LG mothers. Cross-fostering studies have consistently indicted that adult offspring commonly resemble their foster rather than their biological mothers in stress-related behavioral phenotypes and in rearing behavior (Cameron et al., 2008). Levels of hippocampal glucose receptor (GR) are involved in the regulation of the HPA stress response though a negative feedback relationship, with higher levels of GR mRNA associated with lower levels of stress responsivity. Conversely, heightened-GR expressivity is associated with a reduced HPA stress response. Analysis of the of exon 17 glucocorticoid receptor promoter (GR17) in hippocampal neurons several weeks after the birth of pups raised by low- and high-LG mothers reveals significant differences in degree of methylation: High maternal LG is associated with decreased GR17 methylation, corresponding to elevated levels of GR receptor expressivity in the hippocampus and decreased HPA
Political Science and Behavior Genetics
131
stress response (Weaver et al., 2004). Conversely, low-maternal LG is associated with increased GR17 methylation, corresponding to decreased levels of GR receptor expressivity in the hippocampus and increased HPA stress response. In a cross-fostering study (Weaver et al., 2004), the biological offspring of high or low-LG mothers exhibited a pattern of exon 17 methylation associated with the rearing mother, not the biological mother (F ¼ 4.8, Po0.05). In low-LG offspring fostered to high-LG dams, methylation of the 5u CpG dinucleotide site within GR17 in the hippocampus was indistinguishable from that of the biological offspring of high-LG mothers. Conversely, the methylation of the same 5u CpG dinucleotide in the biological offspring of high-LG mothers reared by low-LG dams was comparable to that of low-LG offspring. Strikingly, just before birth at E20, the entire region was unmethylated in both groups. One day after birth-postnatal day P1-the exon 17 GR promoter was de novo methylated in both groups to the same extent. By P6, however, the NGFI-A response element 5u CpG dinucleotide was effectively demethylated in the high-LG, but not the low-LG group, with the differences in methylation remaining consistent through to adulthood. These findings suggest that the group difference in DNA methylation occur as a function of maternal behavior over the first week of life. In the rodent, central oxytocin (OT) receptor levels are functionally linked to behavioral differences in maternal care. ERa expressivity in the MPOA of postnatal day 6 offspring of high-LG-arch-backed nursing (ABN) mothers is significantly increased (t(8) ¼ 4.56, Po0.01) by comparison with the offspring of low-LG-ABN mothers, and levels of methylation at the ERa promoter are significantly elevated in offspring of low, compared with highLG mothers (Champagne & Curley, 2009). In a cross-fostering study, ERa expressivity in the MPOA of adult female offspring revealed a highly significant effect of the rearing, as opposed to the biological mother (F(1,26) ¼ 31.73, Po0.001): ERa expressivity in the MPOA of adult female offspring of low-LG-ABN mothers, but reared by high-LG-ABN dams, was indistinguishable from that of the normal offspring of high-LG-ABN mothers. Similarly, ERa expressivity in the MPOA of biological female offspring of high-LG-ABN mothers reared by low-LG-ABN dams, was indistinguishable from the normal offspring of low-LG mothers. These findings are consistent with both human and primate studies demonstrating the matrilineal transmission of maternal behavior. It should be emphasized in relation to these studies that, as the researchers themselves emphasize, the claim is not that behaviors such as high or low stress reactivity and maternal rearing behavior are in effect caused by
132
EVAN CHARNEY
changes in the methylation status of one or two genes. In studying the epigenetics of the maternal environment, Weaver, Meaney, and Szyf (2006) identified W900 genes as being stably regulated by maternal care. Using microarrays to monitor changes in hippocampal expression of 31,099 unique mRNA transcripts, they identified 253 transcripts that were up-regulated and 50 transcripts that were down-regulated in the offspring of high-LG mothers in comparison with offspring of low-LG dams. Inasmuch as the authors only included genes with known biological function in their analysis, and only studied gene expression in the hippocampus, it is likely that a far greater number of genes are involved. These two studies are intended as representative of numerous studies indicating the manner in which the early rearing environment, and the maternal environment in particular, can reprogram the epigenome with potentially lifelong behavioral consequences. This is the same environment that according to the principle of minimal parental effects (henceforth, the principle of minimal maternal effects (MEs)), has a shared influence on behavioral phenotypes indistinguishable from zero (Plomin & Daniels, 1987).
MATERNAL EFFECTS AND ADAPTATION In evolutionary biology, ME are defined as the effect that occurs when a parent’s phenotype directly affects the phenotype of offspring (Mousseau & Fox, 1998). MEs are ubiquitous in evolutionary biology and ecology, and are particularly prominent in mammals. Mammals are unique among vertebrates in that mothers and offspring have an intimate and extended association during both gestation and lactation (the name of the taxon ‘‘mammalia’’ is derived from the mammary glands of females). The extended period of maternal care and offspring dependence characteristic of many mammals provides a period during which maternal behavior can significantly influence the behavior of offspring (Fusco & Minelli, 2010). ME comprise an important component of phenotypic plasticity in the developing organism. Phenotypic plasticity can be defined broadly as the ability of an organism to change phenotype in response to its environment. This includes the possibility of modifying developmental trajectories in response to specific environmental cues, and the ability of an individual organism to change its phenotypic state or activity in response to variations in environmental conditions. Modern evolutionary biology reflects the idea that adaptation is not limited to the process of natural selection
Political Science and Behavior Genetics
133
(i.e., adaptation at the level of the species), but includes adaptation of the individual organism to its ecological niche. Offspring do not inherit simply genes from their parents, but also an environment. Developmental plasticity evolved because it is adaptive, promoting Darwinian fitness by enhancement of survival and reproductive success by using environmental cues to optimize the life-course strategy (Gluckman et al., 2009). Adaptive phenotypic plasticity is apparent in plants, insects, reptiles, birds, and mammals. What unites almost all species is the centrality of ME as a mechanism of phenotypic plasticity (Linksvayer & Wade, 2005). Offspring behavioral plasticity enables the mother to adjust the phenotype of offspring in response to the environment she inhabits, and in doing so, in effect transmit to them the information about the environment they will inhabit. If the mother’s adjustments to the environment are adaptive, and if the environment is stable across generations, that is, if the cues from the mother’s environment are a good predictor of the environment in which offspring will find themselves, then the offspring’s phenotypic adjustments are adaptive. For example, various studies in humans have shown that the diet of the pregnant mother and/or the nursing mother can significantly influence a child’s food preferences to match the food the mother consumes (Hausner, Nicklaus, Issanchou, Molgaard, & Moller, 2010). As described above, cross-fostering studies of rodents provide strong evidence for ME upon a range of behaviors in offspring, including stress and fear response. Adversity during perinatal development could be expected to forecast an increased level of demand in adulthood. Under conditions of increased environmental demand, the animal’s best interest is to enhance its behavioral (e.g., vigilance, fearfulness) and endocrine (HPA and metabolic) responsivity to stress. These responses promote detection of potential threat, avoidance learning, and metabolic/cardiovascular responses that are essential under the increased demands of the stressor. In regard to maternal behavior, under high-risk conditions, when the probability of extended periods of growth and survival are low, the optimal strategy is to maximize the number of offspring through accelerated mating, increasing the chances that at least some offspring will survive to reproductive maturity (Cameron et al., 2008). Moreover, since adverse environments are characterized by high, unavoidable risks and thus increased mortality, parental investment in offspring quality may be futile. Such conditions favor a shift in reproductive investment toward quantity. In contrast, more favorable environmental conditions favor greater investment in individual offspring at the cost of mating, since offspring quality predicts successful competition for available resources and reproductive fitness. Thus, the quality of the prevailing
134
EVAN CHARNEY
environment defines parental investment, which in turn, is reflected in the mating and parental behavior of the offspring. Were it true that there were no ME upon the behavioral psychology of human offspring, then humans would indeed be exceptional among all animal taxa. What this would entail is that humans possessed less behavioral phenotypic plasticity, were less capable of being behaviorally shaped by ME, than rodents. Any assumption of the absence of ME in this way, both preand postnatally, seems doubly counterintuitive given, as noted, that ME arguably play a larger role in the evolutionary dynamics and adaptation of mammals than in any other animal taxa (Fusco & Minelli, 2010). As mammals, humans have extended and intimate maternal associations, both pre- and postnatal, lasting many years. Yet, in the absence of behavioral ME, what is this commitment for, involving, as it does, an enormous expenditure of maternal time and energy? Just as pressing is the question, what possible adaptive advantage could accrue to the human species by lacking the plasticity to be behaviorally shaped by ME to better meet the demands of its environment? Recent developments in genetics indicate the following: DNA, the ‘‘master molecule,’’ must now share the stage with an ever-growing array of actors, the behavior of which it neither ‘‘controls’’ nor ‘‘determines.’’ Once viewed as a paragon of stability, DNA now appears subject to all manner of structural transformations, some of them, such as retrotransposition, potentially environmentally driven. Instead of an identical DNA molecule in all somatic cells, it appears likely that the normal human condition is one of somatic mosaicism, and instead of one set of chromosomes in each cell, we have chromosomal somatic mosaisicism. All of this variability overturns the simplicity and symmetry of universal, unvarying ratios for shared separating genes based on consanguinity. Pioneering research in epigenetics indicates the manner in which the environment, beginning at the moment of conception and continuing throughout the life course, but particularly in the critical pre- and postnatal period, can profoundly shape behavioral phenotypes by programming the epigenome. Complexity, stochasticity, and environmental responsiveness underlie phenotypic plasticity, and phenotypic plasticity is a key to evolutionary success. That processes such as epigenetic reprogramming, retrotransposition, CSM, and copy-number variation appear to be particularly prevalent in the human brain, that most plastic of organs, is likely no evolutionary accident. In most ways, twin, adoption, and family studies have remained unchanged since the early 20th century. A variant of the twin study was the original biometrical design, proposed by Galton in 1876 (Plomin, 1990).
Political Science and Behavior Genetics
135
The first known ‘‘classical twin study,’’ based upon the observed differences between MZTs and DZTs, was published in 1924, while the first adoption study was proposed by Galton in 1869. That these methodologies have a long pedigree is of course not an argument against them. However, the genetic paradigm on which they are based has also remained largely unchanged. Scientific progress, however, is rarely a matter of accumulating ever more data to be interpreted according to the principles of a paradigm that remains unchanged through the centuries. One might object to this by pointing to advances in statistical analysis. An overreliance upon statistics, however, without paying sufficient attention to developments related to the underlying genetic and biological processes they are supposed to be explaining, leads to unsound conclusions. Statistical methods are sound only to the extent that the scientific paradigm that they are supposed to be elaborating – and which they simultaneously presuppose – is sound. It is time for political scientists attracted by the methods of behavior genetics (BG) to take a careful look at this paradigm in the light of recent developments in the science of genetics, lest they end up explaining political behavior by the principles of 19th century science.
NOTE 1. For a discussion of the epigenome and DNA methylation, see in setion ‘‘The epigenome.’’
REFERENCES Beraldi, R., Pittoggi, C., Sciamanna, I., Mattei, E., & Spadafora, C. (2006). Expression of line-1 retroposons is essential for murine preimplantation development. Molecular Reproduction and Development, 73(3), 279–287. Brouha, B. (2003). Hot L1s account for the bulk of retrotransposition in the human population. Proceedings of the National Academy of Sciences, USA, 100, 5280–5285. Bruder, C. E. G. (2008). Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. American Journal of Human Genetics, 2, 1–9. Callinan, P. A., & Batzer, M. A. (2006). Retrotransposable elements and human disease. In: J. N. Volff (Ed.), Genome and disease (pp. 105–115). Basel, Switzerland: Karger. Cameron, N., Del Corpo, A., Diorio, J., McAllister, K., Sharma, S., & Meaney, M. J. (2008). Maternal programming of sexual behavior and hypothalamic-pituitary-gonadal function in the female rat. PLoS One, 3(5), e2210.
136
EVAN CHARNEY
Carrel, L., & Willard, H. F. (2005). X-inactivation profile reveals extensive variability in X-linked gene expression in females. Nature, 434(7031), 400–404. Champagne, F. A., & Curley, J. P. (2009). Epigenetic mechanisms mediating the long-term effects of maternal care on development. Neuroscience & Biobehavioral Reviews, 33(4), 593–600. Clay Montier, L. L., Deng, J. J., & Bai, Y. (2009). Number matters: Control of mammalian mitochondrial DNA copy number. Journal of Genetics and Genomics, 36(3), 125–131. Coufal, N. G., Garcia-Perez, J. L., Peng, G. E., Yeo, G. W., Mu, Y., Lovci, M. T., et al. (2009). L1 retrotransposition in human neural progenitor cells. Nature, 460(7259), 1127–1131. Dawkins, R. (2006). The selfish gene. New York: Oxford University Press. Dear, P. H. (2009). Copy-number variation: The end of the human genome? Trends in Biotechnology, 27(8), 448–454. Fanciulli, M., Petretto, E., & Aitman, T. (2010). Gene copy number variation and common human disease. Clinical Genetics, 77(3), 201–213. Farthing, C. R., Ficz, G., Ng, R. K., Chan, C. F., Andrews, S., Dean, W., et al. (2008). Global mapping of DNA methylation in mouse promoters reveals epigenetic reprogramming of pluripotency genes. PLoS Genetics, 4(6), e1000116. Faulkner, G. J., Kimura, Y., Daub, C. O., Wani, S., Plessy, C., Irvine, K. M., et al. (2009). The regulated retrotransposon transcriptome of mammalian cells. Nature Genetics, 41(5), 563–571. Flanagan, J. M., Popendikyte, V., Pozdniakovaite, N., Sobolev, M., Assadzadeh, A., Schumacher, A., et al. (2006). Intra- and interindividual epigenetic variation in human germ cells. The American Journal of Human Genetics, 79(1), 67–84. Fowler, J. H., & Dawes, C. T. (2008). Two genes predict voter turnout. Journal of Politics, 70(3), 579–594. Fraga, M. F., Ballestar, E., Paz, M. F., Ropero, S., Setien, F., Ballestar, M. L., et al. (2005). Epigenetic differences arise during the lifetime of monozygotic twins. Proceedings of the National Academy of Sciences of the United States of America, 102(30), 10604–10609. Fusco, G., & Minelli, A. (2010). Phenotypic plasticity in development and evolution: Facts and concepts. Introduction. Philosophical Transactions of the Royal Society of London B, 365(1540), 547–556. Gibbs, R. A. (2003). The international HapMap project. Nature, 426(6968), 789–796. Gluckman, P. D., Hanson, M. A., Bateson, P., Beedle, A. S., Law, C. M., Bhutta, Z. A., et al. (2009). Towards a new developmental synthesis: Adaptive developmental plasticity and human disease. The Lancet, 373(9675), 1654–1657. Hausner, H., Nicklaus, S., Issanchou, S., Molgaard, C., & Moller, P. (2010). Breastfeeding facilitates acceptance of a novel dietary flavour compound. Clinical Nutrition, 29(1), 141–148. He, Y., Wu, J., Dressman, D. C., Iacobuzio-Donahue, C., Markowitz, S. D., Velculescu, V. E., et al. (2010). Heteroplasmic mitochondrial DNA mutations in normal and tumour cells. Nature, 464(7288), 610–614. Holliday, R. (1994). Epigenetics: An overview. Developmental Genetics, 15(6), 453–457. Huang, C. R. L., Schneider, A. M., Lu, Y., Niranjan, T., Shen, P., Robinson, M. A., et al. (2010). Mobile interspersed repeats are major structural variants in the human genome. Cell, 141(7), 1171–1182. Jablonka, E., & Raz, G. (2009). Transgenerational epigenetic inheritance: Prevalence, mechanisms, and implications for the study of heredity and evolution. The Quarterly Review of Biology, 84(2), 131–176.
Political Science and Behavior Genetics
137
Jirtle, R. L., & Skinner, M. K. (2007). Environmental epigenomics and disease susceptibility. Nature Reviews Genetics, 8(4), 253–262. Kaminsky, Z. A., Tang, T., Wang, S.-C., Ptak, C., Oh, G. H. T., Wong, A. H. C., et al. (2009). DNA methylation profiles in monozygotic and dizygotic twins. Nature Genetics, 41(2), 240–245. Kano, H., Godoy, I., Courtney, C., Vetter, M. R., Gerton, G. L., Ostertag, E. M., & Kazazian, H. H. (2009). L1 retrotransposition occurs mainly in embryogenesis and creates somatic mosaicism. Genes & Development, 23(11), 1303–1312. Kingsbury, M. A. (2005). Aneuploid neurons are functionally active and integrated into brain circuitry. Proceedings of the National Academy of Science, USA, 102, 6143–6147. Kruyer, H., Mila, M., Glover, G., Carbonell, P., Ballesta, F., & Estivill, X. (1994). Fragile X syndrome and the (CGG)n mutation: Two families with discordant MZ twins. American Journal of Human Genetics, 54(3), 437–442. Liang, Q., Conte, N., Skarnes, W. C., & Bradley, A. (2008). Extensive genomic copy number variation in embryonic stem cells. Proceedings of the National Academy of Sciences, 105(45), 17453–17456. Linksvayer, T. A., & Wade, M. J. (2005). The evolutionary origin and elaboration of sociality in the aculeate hymenoptera: Maternal effects, sib-social effects, and heterochrony. The Quarterly Review of Biology, 80(3), 317–336. Luo, M., Cui, X., Fredman, D., Brookes, A. J., Azaro, M. A., Greenawalt, D. M., et al. (2009). Genetic structures of copy number variants revealed by genotyping single sperm. PLoS One, 4(4), e5236. Lynch, M., Koskella, B., & Schaak, S. (2006). Mutation pressure and the evolution of organelle genomic architecture. Science, 311(5768), 1727–1730. Marchetto, M. C. N., Gage, F. H., & Muotri, A. R. (2010). Retrotransposition and neuronal diversity. In: H. Ulrich (Ed.), Perspectives of stem cells: From tools for studying mechanisms of neuronal differentiation towards therapy (pp. 87–96). Netherlands: Springer. Martin, S. L. (2009). Developmental biology: Jumping-gene roulette. Nature, 460(7259), 1087–1088. Matlik, K., Redik, K., & Speek, M. (2006). L1 antisense promoter drives tissue-specific expression of human genes. Journal of Biomedicine Biotechnology, 1, 71753. Mkrtchyan, H., Gross, M., Hinreiner, S., Polytiko, A., Manvelyan, M., Mrasek, K., et al. (2010). Early embryonic chromosome instability results in stable mosaic pattern in human tissues. PLoS One, 5(3), e9591. Modi, D., Berde, P., & Bhartiya, D. (2003). Down syndrome: A study of chromosomal mosaicism. Reproductive Biomedicine Online, 6(4), 499–503. Mousseau, T. A., & Fox, C. W. (1998). Maternal effects as adaptations. New York: Oxford University Press. Muotri, A. R., Zhao, C., Marchetto, M. C. N., & Gage, F. H. (2009). Environmental influence on L1 retrotransposons in the adult hippocampus. Hippocampus, 19(10), 1002–1007. Neves, G., Cooke, S. F., & Bliss, T. V. P. (2008). Synaptic plasticity, memory and the hippocampus: A neural network approach to causality. Nature Reviews Neuroscience, 9(1), 65–75. Ostertag, E. M., & Kazazian, H. H. (2001). Biology of mammalian L1 retrotransposons. Annual Review of Genetics, 35, 501–538. Plomin, R. (1990). The role of genetics in behavior. Science, 248(4952), 183–188. Plomin, R., & Daniels, D. (1987). Why are children in the same family so different? Brain and Behavioral Sciences, 10, 1–16.
138
EVAN CHARNEY
Rehen, S. K. (2005). Constitutional aneuploidy in the normal human brain. Journal of Neuroscience, 25, 2176–2180. Rodrı´ guez-Santiago, B., Malats, N. R., Rothman, N., Armengol, L., Garcia-Closas, M., Kogevinas, M., et al. (2010). Mosaic uniparental disomies and aneuploidies as large structural variants of the human genome. The American Journal of Human Genetics, 87(1), 129–138. Seki, Y., Hayashi, K., Itoh, K., Mizugaki, M., Saitou, M., & Matsui, Y. (2005). Extensive and orderly reprogramming of genome-wide chromatin modifications associated with specification and early development of germ cells in mice. Developmental Biology, 278(2), 440–458. Sweatt, J. D. (2009). Experience-dependent epigenetic modifications in the central nervous system. Biological Psychiatry, 65(3), 191–197. Szyf, M., Weaver, I., Champagne, F. A., Diorio, J., & Meaney, M. J. (2005). Maternal programming of steroid receptor expression and phenotype through DNA methylation in the rat. Frontiers in Neuroendocrinology, 26(139). Taylor, R. W., & Turnbull, D. M. (2005). Mitochondrial DNA mutations in human disease. Nature Reviews Genetics, 6(5), 389–402. Trejo, V., Derom, C., Vlietinck, R., Ollier, W., Silman, A., Ebers, G., et al. (1994). X-chromsome inactivation patterns correlate with fetal-placental anatomy in monozygotic twin pairs: Implications for immune relatedness and concordance for autoimmunity. Molecular Medicine, 1(1), 62–70. Weaver, I. C. G., Meaney, M. J., & Szyf, M. (2006). Maternal care effects on the hippocampal transcriptome and anxiety-mediated behaviors in the offspring that are reversible in adulthood. Proceedings of the National Academy of Sciences, USA, 103(9), 3480–3485. Wong, C. C. Y., Caspi, A., Williams, B., Craig, I. W., Houts, R., Ambler, A., et al. (2010). A longitudinal study of epigenetic variation in twins. Epigenetics, 5(6). Yurov, Y. B., Iourov, I. Y., Vorsanova, S. G., Liehr, T., Kolotii, A. D., Kutsev, S. I., et al. (2007). Aneuploidy and confined chromosomal mosaicism in the developing human brain. PLoS One, 2(6), e558.
FROM GENES TO POLITICS: BRIDGING THE GAP BETWEEN BIOLOGICAL AND SOCIAL EXPLANATIONS OF POLITICAL BEHAVIOR VIA TWIN STUDIES Rebecca J. Hannagan In his 2006 annual report for the American Political Science Review (APSR), Lee Sigelman (2006) articulated the following: [A]though only time will tell whether it will emerge among the most important articles the APSR has ever published, one article in particular – Alford, Funk, and Hibbing’s ‘‘Are Political Orientations Genetically Transmitted?’’ – has set a new standard for political science in terms of the media attention and public discussion that its publication has provoked.
Without a doubt, the Alford, Funk, and Hibbing article provoked public discussion, media attention, and critique both from within and beyond the discipline. The issues raised by the article are methodological, theoretical, as well as practical. Advocates as well as detractors have fixed on these in varying attempts to grapple with the ‘‘new standard’’ in political science. This chapter first reviews the articles by political scientists utilizing twin studies beginning with the 2005 Alford, Funk, and Hibbing article. The review will then note the methodological, theoretical, and practical issues Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 139–158 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009008
139
140
REBECCA J. HANNAGAN
that arise from this literature with an eye to what I contend is the more compelling philosophical issue at hand – bringing the body back into politics.
FROM GENES TO POLITICS The 2005 APSR article by John Alford, Carolyn Funk, and John Hibbing presented data from the Virginia 30,000 Health & Lifestyle Questionnaire (VA30K), AARP twin studies, and an Australian twin study (ATR) to test their hypothesis that political attitudes are influenced by genetic as well as environmental factors. Political attitudes, they suggested, were expected to be highly heritable and particularly so on issues most correlated with personality. They employed survey responses from the Wilson–Patterson Attitude Inventory to measure political attitudes. To gauge heritability, they utilize the 2:1 genetic ratio between monozygotic (MZ) and dizygotic (DZ) twins. The authors argued that while previous studies in political attitudes had concentrated on measuring the influence of environmental variables, their test added explanatory power by considering heritability (Alford, Funk, & Hibbing, 2005). Their results indicated that political attitudes were, indeed, highly heritable and even overshadowed the influence of parental socialization in explaining the variation in political attitudes. Heritability for the items on the Wilson–Patterson Attitude Inventory ranged from .41 to .18 and not one of the 28 items’ heritability estimate fell in the single digit range. Their conclusion was not that this finding ‘‘set a new standard’’ or should be considered ‘‘among the most important articles,’’ to use Sigelman’s words, but merely that the findings coupled with evidence on assortative mating may help explain the variation in political viewpoints, which to this point had only been explained via environmental influences (Alford et al., 2005). All in all, the assertion of the meaning and impact of their findings was presented with humility. Words like ‘‘heritability,’’ ‘‘assortative mating,’’ and ‘‘genetics’’ conjure up fears of biological determinism and the specter of Social Darwinism. The emotional response to the work, both in the reporting of it in the popular press as well as the stir it created in the discipline, was palpable. Although it has been presented elsewhere more eloquently, it seems utilitarian to briefly present what Alford, Funk, and Hibbing did, in terms of exactly what a ‘‘twin study’’ entails. The procedure, quite basically, compares MZ (frequently but erroneously called identical) and DZ
Biological and Social Explanations of Political Behavior via Twin Studies
141
(fraternal) twins. MZ twins develop from a single egg fertilized by a single sperm, thus sharing nearly 100% of their genetic inheritance. DZ twins develop from two eggs fertilized by two separate sperm, thus sharing roughly 50% of their genetic inheritance – the same as any full siblings. Classical twin design (CTD), or twin studies, estimate heritability (h2) based on twin correlations: h2 ¼ 2ðrMZ rDZÞ where r is the correlation coefficient. The relative contributions of the shared environmental effects: c2 ¼ rMZ h2 and non-shared environmental effects: e 2 ¼ 1 h2 þ c 2 According to this formula, heritability is an estimate for the relative contribution of genetic effects to total phenotypic variance. Following designs from earlier twin studies, Alford, Funk, and Hibbing applied polychoric correlations to the formula above. Challenges to their findings have been presented primarily as challenges to their methodology. Polychoric correlation transformations are, indeed, limited in that the formula does not: (1) allow for model fitting (testing 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; (4) test for male–female differences in the magnitude of variance components estimates; (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 of interest. In short, there are ‘‘better’’ (i.e., more sophisticated) methods to model heritability. Structural equation modeling (SEM) under a maximum-likelihood (ML) framework is a method used to infer the relative importance of the correlations between observed traits of MZ and DZ twins in terms of their underlying genetic and environmental components. SEM/ML addresses the aforementioned shortcomings by: (1) testing for differences in the zygosity groups; (2) modeling those difference if they exist; (3) including OS twins; (4) including other siblings, parents, and any number of different relative types; and (5) model fitting to determine if removing the genetic or social
142
REBECCA J. HANNAGAN
component of a specific trait provides a statistically better model. As this author and Hatemi illustrated in a 2008 Perspectives on Politics article, retesting the Alford, Funk, and Hibbing assumptions using more rigorous method reinforced their findings. Normal science proceeds incrementally, and in the next iteration of twin studies the authors go beyond the CTD using polychoric correlation transformations and employ SEM/ML, nuclear family models, Bayesian models, as well as begin to pinpoint candidate genes for the phenotypes of interest (Hannagan & Hatemi, 2008). Two years following the APSR article, a team including political scientist Peter K. Hatemi published ‘‘The Genetics of Voting: An Australian Twin Study’’ in Behavioral Genetics. The authors posited that traditional theories of vote choice are based on either a psychological attachment to parties or other identity-invoking groups, or the economic model of rational choice and utility maximization. Seeking to challenge the mainstream theories, the authors employed a twin study to illustrate the extent to which environmental and/or genetic factors influence mass political behavior. Like Alford, Funk, and Hibbing, the team used the ATR data as well as a subset of the Wilson–Patterson Attitude Inventory – the Wilson–Patterson Conservatism scale. From the sample, 11,376 individuals responded with voting preferences and 9,053 responded with specific vote choice (Hatemi, Medland, Morley, Heath, & Martin, 2007). Similar to the findings in the 2005 APSR article, correlations on political attitudes were higher for MZ twin pairs than for DZ twin pairs, suggesting heritability. Adding nuance to the prior study, however, the authors examined differences in opposite sex (OS) pairs and same sex (SS) pairs (hereafter, OS and SS). According to the model used, heritability for vote choice in male pairs was null with the common environment accounting for the majority of the variance. Heritability for vote choice in female pairs was .28 with the common environment accounting for .52 of the variance. The authors concluded by noting that reliance on the rational choice model of vote choice would leave a grossly inadequate understanding of mass political behavior (Hatemi et al., 2007). This article embodies the scientific method as applied to political science. Scientific observation is done with an eye to reliability and replication, quantification and hypothesis testing (empirical tests imply falsifiability), and finally generalizability. At this stage in the evolution of the use of twin studies in political science, the focus is largely on improving methods and the approach is incremental. In 2008, five articles employing twin studies were published in political science journals or in major scientific journals by political scientists. The
Biological and Social Explanations of Political Behavior via Twin Studies
143
article, ‘‘Transmission of Attitudes Toward Abortion and Gay Rights: Effects of Genes, Social Learning, and Mate Selection’’ appeared in Behavioral Genetics and was authored by political scientist Hatemi – coauthoring with Lindon J. Eaves.1 The authors used data from the VA30K, but also expanded the model to measure twins, non-twin siblings, parents, and other members of the nuclear family. The primary interest of the authors in this article was to measure the genetic and environmental influences on attitudes relating to homosexuality and abortion (Eaves & Hatemi, 2008). Hatemi and Eaves noted that traditional social scientific approaches focus on the role of parental socialization in explaining variation in political attitudes. They found that MZ twins have more highly correlated beliefs about abortion and homosexuality than do DZ twins. Genetic effects were estimated to account for 50–70% of the variance, whereas unique environmental experiences for 20–30% of the variance, and social learning for just 10–20% of the variance. The authors highlighted the small effect of social learning and the likely underestimated effect of genetic influence due to assortative mating to further illustrate that past studies are incorrect in arguing that social learning is the key contributor to political attitudes (Eaves & Hatemi, 2008). Like Alford, Funk, and Hibbing, the authors here do not make claims beyond the notion that models to this point have been incomplete, and twin studies provide a mechanism by which they can be enhanced. In a related article, a team including Hatemi published ‘‘Social and Genetic Influences of Adolescent Religious Attitudes and Practices’’ in Social Forces. In this article, the authors utilized data from 953 twin pairs, ranging from 11 to 18 years old, residing in North Carolina. The authors found that in contrast to adult studies of the genetic influence on religiosity, environmental factors were the key determinate for adolescent religiosity. On average, genetic influences were found to account for 10% of the variance, shared environment for 50%, and unique environment for 40% of the variance. The study confirms existing claims regarding the role of socialization in the acquisition of religious attitudes (Eaves, Hatemi, PromWomley, & Murrelle, 2008). The team goes on to make the case that the socialization of religious values is important from an evolutionary perspective. Religiously trained adolescents, they argued, are protected against the expression of high-risk behavioral predispositions that could lead to socially damaging outcomes. They concluded the article by arguing that although genetic factors do not play a role in religious behavior among adolescents, genetic factors
144
REBECCA J. HANNAGAN
overshadow socialization once the adolescent leaves the home environment. This article makes great strides in articulating how we ought to think about, and model, the impact of genetic and environmental influences on something as complex as religious or political attitudes. Family influences are likely to be more important during formative years. Parental impact is also likely to wane over the lifespan (Eaves et al., 2008). This is a notion that has been largely absent from theorizing and modeling attitude formation in political science and it also illustrates the point articulated, but perhaps ignored, in the Alford, Funk, and Hibbing article. The point being that just as environments are not fixed, neither are genes. Beginning with CTD, expanding to nuclear family models, and so on, provides an opportunity to not only identify the relative impact of genetic inheritance but also where environments and what environments have an impact on the attitudes and behaviors of interest to political scientists. In a shift away from examining political attitudes that began with Alford, Funk, and Hibbing and continued with work by Hatemi and colleagues, Fowler, Baker, and Dawes examined political behavior in ‘‘Genetic Variation in Political Participation’’ published in the American Political Science Review in 2008. Fowler, Baker, and Dawes focused on voter turnout, but their premise was the same as that of the previous authors. Instead of focusing on transmitted social norms via parental teaching and socialization, they hypothesized that genetic factors are also at work. In this study, Los Angeles County voter registration records were cross-checked with the Southern California Twin Registry to provide 878 SS twin pairs, of those 535 were MZ and 343 were DZ. Using a Bayesian model the authors found that 53% of the variance in turnout behavior could be accounted for by genetic influence (Fowler, Baker, & Dawes, 2008). In yet another shift from the previous authors’ approaches, Fowler, Baker, and Dawes replicated their results using data from The National Longitudinal Study of Adolescent Health (Add Health) – a much larger and less regional sample. The replication using a different sample corroborated their prior results. In the discussion section of their article the authors argued that while there is evidence of heritability for participation, the link between genes and phenotypes has yet to be pinpointed. They suggest it may be related to other prosocial behaviors since voting may impart satisfaction by fulfilling one’s civic duty (Fowler et al., 2008). Making good on their speculation in the 2008 APSR article, Fowler and Dawes published ‘‘Two Genes Predict Voter Turnout’’ in The Journal of Politics. Two genes related to neurochemical transcription, MAOA and 5HTT, were hypothesized to account for some variation in political
Biological and Social Explanations of Political Behavior via Twin Studies
145
behavior. Fowler and Dawes obtained data from Wave III of Add Health sibling-pairs – including twins, half siblings, and unrelated pairs. Participants provided DNA samples in addition to reporting recent religious and political activities and the authors hypothesized an association between MAOA, 5HTT and voting behavior may be moderated by social activity. Their suggestion is premised on the idea that because these genes affect prosocial behavior via neurotransmitters, they will also affect political participation (Fowler & Dawes, 2008). Alford, Funk, and Hibbing dedicated a section of their 2005 article to a discussion of 5HTT and its likelihood as a candidate for further exploration as well. The genes under consideration, MAOA and 5HTT, have been previously associated with antisocial behavior in studies unrelated to specific political behavior. MAOA and 5HTT manage the reabsorption of excess serotonin into the cell during time of social stress. Both the MAOA and 5HTT polymorphisms are present in ‘‘high/low’’ and ‘‘long/short’’ alleles, respectively. A high MAOA polymorphism and a long 5HTT polymorphism better regulate stress and individuals possessing these alleles were more likely to be expected to vote, according to Fowler and Dawes. Individuals possessing the low and short alleles would be less inclined to vote or to generally avoid activities where there was a significant chance of loss and ensuing stress (Fowler & Dawes, 2008). The results showed that only a ‘‘high’’ version of MAOA was associated with voter turnout – increasing the likelihood by 1.26 times over the ‘‘low’’ polymorphic allele. MAOA was not moderated by religious attendance whereas 5HTT was. Those possessing the long version of 5HTT who participated in religious activities were 1.58 times more likely to vote. Those possessing the short allele, even if they participated in religious activities, were not more likely to vote as they are less able to mitigate the effect of a possible loss, according to the authors (Fowler & Dawes, 2008). This article illustrates gene and environment interaction, and though previous studies noted the importance of thinking about the interaction as opposed to a falsely dichotomous ‘‘nature versus nurture’’ way of thinking, this is the first time it has been illustrated in an empirical test. The authors show that both the long version of 5HTT and the environmental factor – participation in religious activities – were necessary to explain the increase in likelihood of voting as opposed to merely looking to the presence of the MAOA polymorphism or merely looking to religious participation. In a related article appearing in PNAS in 2009, Rose McDermott and coauthors found that the ‘‘low’’ version of MAOA significantly predicts punishing behavior in an experimental setting where subjects are provoked
146
REBECCA J. HANNAGAN
by unfair behavior. In ‘‘Monoamine oxidase A gene (MAOA) predicts behavioral aggression following provocation,’’ the authors note the importance of evidence for genetic influence on aggression and punishing behavior as well as the greater theoretical importance of considering the nuance of everyday behaviors and strategies (McDermott, Tingley, Cowden, Frazzetto, & Johnson, 2009). Fowler and Dawes, this time with a new team of coauthors, published ‘‘Heritability of cooperative behavior in the trust game’’ in PNAS. The team utilized data on 658 MZ and DZ SS twins from the Swedish Twin Registry, as well as 706 twins from the 2006 and 2007 Twin Days Festival. Their subjects participated in a trust game with real money between anonymous partners where an investor gave an endowment to a trustee. The money was then multiplied and the trustee had the option to return an amount to the investor. The game thus measures both trust and trustworthiness (Cesarini et al., 2008). Individual differences in trust and trustworthiness due to common environment, unique environment, and genetic influence were estimated using a mixed-effects Bayesian analysis. Genetic variation was calculated by examining the difference in correlation between MZ and DZ twins, as in previously cited studies. Behavior of MZ twins was more consistent than that of DZ twins suggesting genetic influences on both trust and trustworthiness. Genetic differences were found to be more influential than variances influenced by the common environment but less than those of the unique environment (Cesarini et al., 2008). In his 2008 Science article with Darren Schreiber, Fowler argues for political scientists and biologists to work together as genetic variation is increasingly shown to play a role in political behaviors. This article, ‘‘Biology, Politics, and the Emerging Science of Human Nature,’’ does not employ a twin study but suggests that such studies have clearly established that both genes and the environment play a role in the development of political attitudes and behaviors. The next steps, according to Fowler and Schreiber, are to pinpoint specific genes as well as networks of brain regions active during political thought, that give rise to political behaviors (Fowler & Schreiber, 2008). Though a brief note, this article is a call to duty, of sorts, and articulates some of the same themes this author and Hatemi cited in the Perspectives on Politics article. In short, the authors argue that this is work that should be done by political scientists. If we instead choose to wrangle about the problems with twin studies, we will cede a large part of our discipline to those in psychology and behavioral genetics who are already working on topics of interest to our discipline (Hannagan & Hatemi, 2008).
Biological and Social Explanations of Political Behavior via Twin Studies
147
The practical matter associated with political scientists and biologists working together is an issue I take up in the following section. ‘‘Is There a ‘Party’ in your Genes?’’ was published in Political Research Quarterly in 2009. Hibbing and Alford, joined by Hatemi, Eaves, and Martin, examined the sources of party identification as well as the intensity of that identification. Data were obtained from the VA30K as well as the Mid-Atlantic Twin Registry. Their results showed that genes exert little, if any, influence on party identification but that they appear to play a critical role in shaping the strength of a person’s party identification (Hatemi, Alford, Hibbing, Martin, & Eaves, 2009). One of the important contributions of this article is the care taken to examine the link between the heritability of personality and the likely importance for political attitudes. This article is the foundation for more ‘‘mainstream’’ articles to come such as Mondak and colleagues’ 2010 article ‘‘Personality and Civic Engagement: An Integrative Framework for the Study of Trait Effects on Political Behavior’’ (Mondak, Hibbing, Canache, Seligson, & Anderson, 2010). The Journal of Politics published Dawes and Fowler’s ‘‘Partisanship, Voting, and the Dopamine D2 Receptor Gene’’ in 2009. The authors hypothesized that the DRD2 gene influences whether or not an individual will identify with a political party and subsequently whether or not they will turn out to vote. More specifically, individuals possessing one or two copies of the A2 allele would be more likely to engage in partisan behavior. Data for this study were derived from 90,114 adolescent siblings representative of the general population from Add Health. The A2 allele was found to have a significant association with partisanship even with control variables (i.e., age, gender, race, and other socioeconomic indicators) held at means (Dawes & Fowler, 2009). The study was then replicated with 91 sibships carrying varied DRD2 genotypes and again confirmed the association while ruling out any possible distortion due to population stratification. The results of these studies led the authors to contend that the ability to form social attachments and/or improved cognitive function, which has been previously associated with the A2 allele, increases the likelihood that an individual will form an attachment with a political party. These results may also explain the strong tendency for partisan behavior to run in families give the genetic interaction with parental socialization (Dawes & Fowler, 2009). Settle, Dawes, and Fowler published ‘‘The Heritability of Partisan Attachment’’ in Political Research Quarterly. Like the Dawes and Fowler JOP article, the authors contend that party identification can no longer be conceived of as merely a result of socialization. They examined influences on
148
REBECCA J. HANNAGAN
the strength of partisan attachment, specifically whether heritability plays a demonstrative role. They contended party identification is likely tied to environmental influences, but that the intensity of one’s partisan attachment will be influenced more by genetics than socialization. Data were derived from 353 SS twin pairs from the 2006 and 2007 Twin Days Festival. Participants completed a questionnaire that included demographic questions as well as one concerning party affiliation (Settle, Dawes, & Fowler, 2009). The results suggested that heritability accounts for 46% of the variance in the strength of partisan attachment. Unshared environmental factors were also shown to account for 54% of the variance. As with previous studies, no significant role for heritability in determining party affiliation was found, but once again the case for a genetic role in the disposition for political involvement was substantiated (Settle et al., 2009). One of the obvious contributions stemming from the work done by Fowler, Dawes, and coauthors is replication with different sample populations. When one begins to see the threads through all these studies, it is increasingly clear that there is more to this literature than merely illustrating that ‘‘environment does not explain all the variance’’ though that is the consistent message articulated. Fowler and Dawes, this time working with Nicholas A. Christakis, published ‘‘Model of Genetic Variation in Human Social Networks’’ in PNAS in 2009. The authors hypothesized that genes play a role in human social networks structures and tested this idea using the twin study design. Data were derived from 1,100 twins from Add Health. Results showed genetic factors accounted for 46% of variation in in-degree (how many times a person is named as a friend), 22% of out-degree (how many friends a person names), and 29% of between-ness centrality (network centrality). The authors were surprised that in-degree was found to be heritable, whereas out-degree was not. They also noted that heterogeneity among individual nodes is a necessary prerequisite for heritability in a model of human social networks. Traits of individual nodes are thought to be just as important as structural processes in understanding networks (Fowler, Dawes, & Christakis, 2009). The authors also surmise that reasons for genetic variation in social networks include the augmentation of social support and group solidarity though they concede that the proximate causes for variation are likely linked to natural selection working on other traits (Fowler et al., 2009). This article fits with the overall trend in the Fowler and Dawes contribution to the literature: that prosociality – aspects of which are known to be genetically linked – play out through trust, social networks, group participation, and, thus, are likely to also be related to strength of group-based conviction and political participation.
Biological and Social Explanations of Political Behavior via Twin Studies
149
A related article appearing in The Quarterly Journal of Economics, by Dawes and coauthors, utilized the CTD to provide estimates of environmental and genetic influence on preferences for risk taking behavior and giving behavior in an experimental setting. The authors found evidence that genetic differences explain approximately 20% of variation in individual preferences for this political behavior (Cesarini, Dawes, Johannesson, Lichtenstein, & Wallace, 2009). Although the authors published the article in an economics journal, the application of risk orientation and subsequent behavior is important for political scientists. Coupled with the findings from McDermott and colleagues, the literature illustrates applicability to political psychology broadly construed, as opposed to merely focusing on attitudes of interest to scholars of American political behavior. In an article by Hatemi and Eaves, this time coauthoring with Sarah Medland, published in the JOP, the authors ask ‘‘Do Genes Contribute to the ‘Gender Gap’?’’ The authors hypothesized that if genetic dispositions are a result of the adaptive responses of humans throughout evolutionary history, we might expect sex differences in the magnitude of genetic influences on political attitudes. Data were derived from the VA30K where 70% of 14,763 invited twin pairs returned completed questionnaires. The authors found evidence that there is variation in the genetic influence on political attitudes between men and women for certain issues – namely divorce. The authors concluded that modern political preferences related to sexually dimorphic adaptive traits are likely the best candidates for identifying different sources of genetic influence (Hatemi, Medland, & Eaves, 2009). The latest article, published in 2010 by the JOP, is by Settle, Dawes, Christakis, and Fowler. The work is titled ‘‘Friendships Moderate an Association between a Dopamine Gene Variant and Political Ideology,’’ and here, the authors seek to understand more complex political behaviors using a polygenic approach – one that examines specific gene–environment interactions. More specifically, the authors examine the 7R allele of the dopamine D4 receptor gene (DRD4) and its interaction with the social variable of how many friendships a person forms. The 7R allele of DRD4 is known to be associated with novelty-seeking and sensation-seeking, openness to new experiences and other behaviors correlated with people of liberal political ideology. Data is derived from Add Health with 2,574 individuals providing genetic markers identifying DRD4 alleles. Results show that simply having the 7R allele or having many childhood friends does not increase the tendency toward liberal thought. The results did demonstrate, however, that having more 7R alleles and more friends during adolescence is associated with being more liberal (Settle, Dawes, Christakis, & Fowler, 2010).
150
REBECCA J. HANNAGAN
One of the ways to make sense of the literature to date is to treat it as descriptive data. There are essentially two main camps producing this literature: The Hibbing/Nebraska group and the Fowler/University of California at San Diego (hereafter, UCSD) group. Although Hatemi and Dawes have certainly produced a number of the publications in their own right, they are the former graduate students of Hibbing and Fowler, respectively, and thus for simplicity can be considered part of those camps. A nearly equal number of publications have come from the two groups and have been published in many of the same journals. Both groups have published roughly half of their work in mainstream political science journals and the rest in scientific journals or journals in other disciplines. The Nebraska group has focused more on methodological techniques as well as the genetic basis for attitudes that can explain differences in political ideology, whereas the UCSD group has focused more on replication of studies and creating a coherent body of literature around the idea that variation in prosocial behavior is linked to specific genes and likely underpins political behaviors such as voting and joining a political party as well as strength in partisan attachment.
Authors
Publications (cited here) Journals (cited here)
Nebraska Group
UCSD Group
John R. Hibbing, John R. Alford, Carolyn Funk, Kevin B. Smith, Peter K. Hatemi, Lindon J. Eaves, Nicholas G. Martin, Sarah Medland, Katherine Morley, Andrew Heath, Lenn Murrelle, Elizabeth PromWomley, Matthew Keller, Laramie Duncan, Michael Neale, Hermine Maes 9
James H. Fowler, Christopher Dawes, Laura Baker, Darren Schreiber, Jamie Settle, Nicholas Christakis, David Cesarini, Magnus Johannesson, Paul € Wallace Lichtenstein, Bjorn
APSR, JOP, AJPS, Political Research Quarterly, Political Analysis, Behavioral Genetics (2), Social Forces, Twin Research and Human Genetics
APSR, JOP (3), Political Research Quarterly, Quarterly Journal of Economics, PNAS (2), Science
9
Biological and Social Explanations of Political Behavior via Twin Studies
151
Nebraska Group
UCSD Group
Data sets
VA30K, AARP, ATR, North Carolina twin study, MidAtlantic Twin Registry
Design/ methods
CTD, SEM/ML, ACE/CE models, extended family models
Southern CA Twin Registry, Add Health, 2006 and 2007 Twin Days Festival, Swedish Twin Registry CTD, Bayesian ACE/CE models
Candidate genes identified
MAOA, 5HTT, DRD2, DRD4
Note that the number of publications and journals in which they were published is limited to the sample considered for this review. There were a number of articles published by both groups in nonpolitical science journals where the research question was far enough removed from the interest of mainstream political science to be eliminated from the sample. That was simply this author’s choice, but the rationale is that if the articles selected are representative of the work produced since the 2005 Alford, Funk, and Hibbing article, they are sufficient to say something about the scope and impact of the entire literature produced. There are some practical considerations that arise from this body of work. That the majority of this literature has come from two groups implies there is something special about either UCSD and The University of Nebraska or James Fowler and John Hibbing. Although I am sure that is true in many respects, both have argued for engagement in this work by a plurality of scholars. Hatemi and colleagues have published numerous articles detailing the methods employed (i.e., Keller et al., 2008; Medland & Hatemi, 2009; Hatemi et al., 2010). Hatemi has also hosted National Science Foundation sponsored workshops on methods – bringing behavioral geneticists, psychologists, and political scientists together. Although there is a reasonable assumption that the data used in these studies is proprietary, Fowler and Dawes illustrate that much of the data they use (e.g., Add Health) is available to anyone. There is also the practical consideration involved in actually obtaining advanced methods training and learning the literature associated with behavioral genetics. Most political scientists are narrowly trained in graduate school, narrowly focus on their area of expertise as an assistant professor to obtain tenure, and as associate professors may be reluctant to engage in a new approach that may counter their prior work. In short, the disciplinary barriers are high and institutional practices tied to tenure
152
REBECCA J. HANNAGAN
largely preclude innovative and creative work that requires going outside the discipline. As demonstrated by the list of coauthors on many of the articles cited earlier, coauthoring is a strategy not only for division of labor purposes but also because collaborative research can be much more rewarding and interesting if collaborators are outside the discipline of political science. Even if it is impractical to engage in collaborations with biologists, there are many political scientists who either have training in this area or are interested in learning new approaches. There is good reason to be encouraged by the work done so far by Hibbing, Fowler, Hatemi, Dawes, and so on, and the journals in political science are now accustomed to seeing manuscripts employing these methodologies. Many graduate students and junior faculty attend ICPSR methods training in the summer and workshops to teach methods used in behavioral genetics are just as accessible. Learning new tools is something nonthreatening to many political scientists, but to others this methodology is still considered quite distinct from a summer ICPSR workshop on game theory. So, what is the rub? Critics of the 2005 APSR article centered mainly on the methodological questions. ‘‘What is a heritability estimate? Is it meaningful? Is the Equal Environment Assumption (EEA) problematic?’’ Such issues have been raised and countered by Alford, Funk, and Hibbing and expanded upon by Hatemi and colleagues in a number of published articles. For those fundamentally troubled by the suggestion that political attitudes are the result of more than socialization, the introduction to new methods and subsequent improvements on methodology offer little satisfaction. Academic dissent often begins with the low-hanging fruit of methodology. ‘‘This cannot be so, so I will illustrate that it was done incorrectly, or, that it cannot be done correctly, etc.’’ And as unsatisfying as retorts to methodological critiques are to the detractors, such are the critiques to the advocates in the first place. ‘‘Of course we can estimate heritability, and it is no less meaningful than a regression coefficient, etc. The EEA is not problematic and has been addressed ad nauseam, etc.’’ But this is not fundamentally a debate between social constructionist naysayers and ‘‘greedy’’ reductionists – though that may appear to be the case as presented in the political science forums to date. It is perfectly acceptable to discuss the virtues of using SEM over CTD, but such issues are not the reason for this literature garnering significant media attention nor its evocation of strong emotions within the discipline. This is about learning neither new methods nor certain methods being preferenced over others. This is not a Perestroikan revival in the making. The real issue behind the media attention and discomfort experienced by some at the
Biological and Social Explanations of Political Behavior via Twin Studies
153
mention of ‘‘genes,’’ and the truly thought-provoking aspect of the Alford, Funk, and Hibbing article and the subsequent literature utilizing twin studies, is the notion that we are biological.
BRIDGING THE GAP BETWEEN BIOLOGICAL AND SOCIAL EXPLANATIONS This idea – that we are biological – has been lacking (almost) entirely in our theorizing about and empirical tests of political behavior. When E. O. Wilson published the infamous Sociobiology in 1976 he suggested that the social sciences would inevitably converge on the biological sciences. Misinterpretation of Wilson’s aims gave rise to more tension between the biological and social constructionist camps at that time. Sandra Harding once suggested that history has shown us that ‘‘science’’ sometimes means the assumptions, observations, and discoveries of marginalized groups – such as the materialists that challenged the early Church and its reigning paradigm – but other times ‘‘science’’ is the assumptions [of] those y dominant groups, as in the racial, sexist and class-bound biological determinist ones shaping nineteenth century craniology and other studies of intelligence. (Harding, 1995)
It is the view that science is not a mirror of nature, but rather a practice and culture of a particular dominant group that Wilson intended to rectify. Wilson’s objective was (seemingly) not for nefarious ends rooted in ideology, but rather to provide the tools for cross-disciplinary research using an evolutionary framework. Wilson later stated, I expected that many social scientists, already convinced of the necessity of a biological foundation for their subject, would be tempted to pick up the tools and try them out y I now understand that I entirely underestimated the Durkheim-Boas tradition of autonomy of the social sciences, as well as the strength and power of the antigenetic bias that has prevailed as virtual dogma since the fall of Social Darwinism. (Wilson, 1978)
Many scholars have since considered Wilson’s proposal in Sociobiology to be a ‘‘visionary notion that even human culture is itself a biological phenomenon’’ (Gowaty, 1997). Our discipline is still suffering from biophobia or, at the very least, is quite unsure what ‘‘picking up the tools’’ of this new approach would actually entail. The suggestion that our political attitudes come from something beyond parental socialization, or culture, or a social construct, or innate rationality
154
REBECCA J. HANNAGAN
is compelling because it conjures up fears of biological determinism and even social Darwinism. Placing something as seemingly esoteric as political attitudes in the realm of biology is frightening to some, threatening to others, and yet to others it is merely common sense, quite obvious, or even exciting. We have long been suffering the hangover from the soire´e of disembodied reason that was the Enlightenment. Bringing the body ‘‘back’’ has been a philosophical as well as an empirical endeavor for a very long time. It would seem, given the progress made in this regard in other disciplines, that political science should be less averse to such an idea. This has not been our disciplinary experience, however, as disembodied reason ensconced in rational choice theory continues to dominate the study of political behavior. Despite having traveled the arduous road to becoming a ‘‘real’’ science through increasingly complex, rigorous, and sophisticated methodologies, our ability to advance as a discipline is stilted. The discomfort felt in the wake of the twin studies literature is a mixture of aversion to the next iteration of statistical methods masquerading as ‘‘real’’ science in the top journals and an aversion to the specter of biological determinism evoked by the topic of ‘‘genetics.’’ Skepticism of this new literature is not unfounded. The behavioral revolution that led to the scientification of political science has provided few answers to questions that truly matter. We are presently guided by a scientific theory that is not falsifiable. Rational choice has left many of us weary of models, not because there is anything wrong with models, but because they have failed to provide that theoretical breakthrough that ‘‘real’’ science seems to require. The true path to development as a scientific discipline is not one of statistical methods but of theoretical foundations.2 Where do we go from here and what role does the literature utilizing twin studies play? The interplay between biology and society is multifaceted, complex, and difficult to measure. The studies cited earlier illustrate a way in which this interplay can be approached. Some social scientists assert that attitudes and behaviors result from cultural learning, and cultural learning is not biological, but that is incorrect because every aspect of every living thing is biological. It is not the case that some human behaviors are in the ‘‘realm of biology’’ and others not. Attitudes and behaviors can be influenced by human society – as illustrated by the studies in this chapter – but cannot be ‘‘entirely cultural’’ and thus, ‘‘not biological.’’ One concept fundamental to a theory of human political behavior is that people cannot learn to occupy and reinforce social roles, attitudes, and behaviors without the underlying
Biological and Social Explanations of Political Behavior via Twin Studies
155
adaptation for such learning (Thornhill & Palmer, 2000). This notion has not been thoroughly recognized in the study of political behavior. As biological beings the structure and functioning of brains and hormones ‘‘pervade our bodies and our interactions with the world in an all-encompassing way’’ that cannot be theoretically or otherwise separated from who we are – regardless of any widely acknowledged theoretical commitments (Warnke, 2007). The next step is to integrate the findings from cognitive science to our normative questions about human political behavior. These questions are rooted in who we are and how we ought to live, but we should be guided by what empirical science says about what a person is as opposed to some a priori understanding of ‘‘person as disembodied reason’’ or ‘‘rational actor.’’ George Lakoff, in Philosophy in the Flesh, presents his argument for an ‘‘empirically responsible philosophy.’’ This entails a reconsideration of our false views of what a person is. He argues the erroneous conceptions we carry are (1) that there is a mind-independent objective world; (2) universal reason characterizes the rational structure of the world; (3) human reason is the capacity to use some portion of universal reason; (4) we can have objective knowledge of the world via the use of universal reason; (5) since human reason is disembodied, it is separate from all bodily capacities; and (6) universal reason separates human beings from the animals (Lakoff & Johnson, 1999). Such views affect and influence every aspect of our lives from morality to politics, but they are so second nature that we barely notice. This erroneous worldview, stemming from our view of what a person is, is the basis of our current rational choice approach to the study of political behavior. Lakoff argues that embodied reason would entail a new view of what a person is. This is a philosophical view rooted in empirical cognitive science. It entails that (1) our conceptual system is shaped by our perceptual and motor systems; (2) we can only form concepts through the body; (3) basiclevel concepts use perceptual, imaging, and motor systems to characterize our functioning in everyday life; (4) major forms of rational inference are instances of sensorimotor inference; (5) because our ideas are framed in terms of our unconscious embodied conceptual systems, truth and knowledge depend on embodied understanding; and (6) because concepts and reason both derive from the sensorimotor system, the mind is not separate from the body (Lakoff & Johnson, 1999). This approach to what a person is may seem reductionist, but it is not the case that a person is determined as a result of being embodied. If the literature employing twin studies illustrates
156
REBECCA J. HANNAGAN
anything, it is the importance of the environment in addition to genetic influence. The emergence of a self, even if rooted in biology, is still a moral entity capable of choice. There is work being done in political science to shift the disciplinary focus away from disembodied reason and the erroneous philosophy of what a person is. Among the key contributions are McDermott’s (2004) ‘‘The Feeling of Rationality: The Meaning of Neuroscientific Advances for Political Science,’’ Lieberman, Schreiber, and Ochsner’s (2003) ‘‘Is Political Cognition Like Riding a Bicycle? How Cognitive Neuroscience Can Inform Research on Political Thinking,’’ and Orbell, Morikawa, and Allen’s (2002) ‘‘The Evolution of Political Intelligence: Simulation Results.’’ Excellent theoretical work such as that by McDermott, Schreiber, and Orbell are necessary for the broader discipline to make sense of the growing literature using twin studies and genetics to explain variation in political attitudes and behaviors. The space between genes and political behavior is vast, and much of the work to be done to fill that space must come from the cognitive sciences using the theoretical framework of evolutionary biology.
NOTES 1. Lindon J. Eaves is distinguished professor in the Departments of Human Genetics and Psychiatry and director of the Virginia Institute for Psychiatric and Behavioral Genetics at Virginia Commonwealth University. 2. Jon R. Bond makes this argument in his 2007 presidential address to the Southern Political Science Association entitled: ‘‘The Scientification of the Study of Politics: Some Observations on the Behavioral Evolution in Political Science.’’
REFERENCES Alford, J. R., Funk, C. L., & Hibbing, J. R. (2005). Are political orientations genetically transmitted? American Political Science Review, 99(2), 153–167. Cesarini, D., Dawes, C. T., Fowler, J. H., Johannesson, M., Lichtenstein, P., & Wallace, B. (2008). Heritability of cooperative behavior in the trust game. PNAS, 105(10), 3721–3726. Cesarini, D., Dawes, C. T., Johannesson, M., Lichtenstein, P., & Wallace, B. (2009). Genetic variation in preferences for giving and risk taking. The Quarterly Journal of Economics, 124(2), 809–842. Dawes, C., & Fowler, J. H. (2009). Partisanship, voting, and the dopamine D2 receptor gene. The Journal of Politics, 71(3), 1157–1171.
Biological and Social Explanations of Political Behavior via Twin Studies
157
Eaves, L. J., & Hatemi, P. K. (2008). Transmission of attitudes toward abortion and gay rights: Effects of genes, social learning and mate selection. Behavioral Genetics, 38, 247–256. Eaves, L. J., Hatemi, P. K., Prom-Womley, E. C., & Murrelle, L. (2008). Social and genetic influences of adolescent religious attitudes and practices. Social Forces, 86(4), 1621–1646. Fowler, J. H., Baker, L. A., & Dawes, C. T. (2008). Genetic variation in political participation. American Political Science Review, 102(2), 233–248. Fowler, J. H., & Dawes, C. T. (2008). Two genes predict voter turnout. The Journal of Politics, 70(3), 579–594. Fowler, J. H., Dawes, C. T., & Christakis, N. A. (2009). Model of genetic variation in human social networks. PNAS, 106(6), 1720–1724. Fowler, J. H., & Schreiber, D. (2008). Biology, politics, and the emerging science of human nature. Science, 322, 912–914. Gowaty, P. A. (1997). Feminism and evolutionary biology. New York: Chapman. Hannagan, R. J., & Hatemi, P. K. (2008). The threat of genes: A comment on Evan Charney’s ‘genes and ideologies’. Perspectives on Politics, 6, 329–335. Harding, S. (1995). Can feminist thought make economics more objective? Feminist Economics, 1, 7–32. Hatemi, P. K., Alford, J. R., Hibbing, J. R., Martin, N. G., & Eaves, L. J. (2009). Is there a ‘‘party’’ in your genes? Political Research Quarterly, 62(3), 584–600. Hatemi, P. K., Hibbing, J. R., Medland, S. E., Keller, M. C., Alford, J. R., Smith, K. B., Martin, N. G., & Eaves, L. J. (2010). Not by twins alone: Using the extended family design to investigate genetic influence on political beliefs. American Journal of Political Science, 54(3), 798–814. Hatemi, P. K., Medland, S. E., & Eaves, L. J. (2009). Do genes contribute to the ‘‘Gender Gap’’? The Journal of Politics, 71(1), 262–276. Hatemi, P. K., Medland, S. E., Morley, K. I., Heath, A. C., & Martin, N. G. (2007). The genetics of voting: An Australian twin study. Behavioral Genetics, 37, 435–448. Keller, M. C., Medland, S. E., Duncan, L. E., Hatemi, P. K., Neale, M. C., Hermine, H. M., & Lindon, E. J. (2008). Modeling extended twin family data I: Description of the cascade model. Twin Research and Human Genetics, 12(1), 8–18. Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: The embodied mind and its challenge to western thought. New York, NY: Basic Books. Lieberman, M. D., Schreiber, D., & Ochsner, K. N. (2003). Is political cognition like riding a bicycle? How cognitive neuroscience can inform research on political thinking. Political Psychology, 24(4), 681–704. Medland, S. E., & Hatemi, P. K. (2009). Political science, biometric theory, and twin studies: An introduction. Political Analysis, 17(2), 191–214. McDermott, R. (2004). The feeling of rationality: The meaning of neuroscientific advances for political science. Perspectives on Politics, 2, 691–706. McDermott, R., Tingley, D., Cowden, J., Frazzetto, G., & Johnson, D. D. P. (2009). Monoamine oxidase A gene (MAOA) predicts behavioral aggression following Provocation. PNAS, 106(7), 2118–2123. Mondak, J. J., Hibbing, M. V., Canache, D., Seligson, M. A., & Anderson, M. R. (2010). Personality and civic engagement: An integrative framework for the study of trait effects on political behavior. American Political Science Review, 104(1), 85–110. Orbell, J., Morikawa, T., & Allen, N. (2002). The evolution of political intelligence: Simulation results. The British Journal of Political Science, 32(4), 613–639.
158
REBECCA J. HANNAGAN
Settle, J. E., Dawes, C. T., Christakis, N. A., & Fowler, J. H. (2010). Friendships moderate an association between a dopamine gene variant and political ideology. The Journal of Politics, 72(4), 1189–1198. Settle, J. E., Dawes, C. T., & Fowler, J. H. (2009). The heritability of partisan attachment. Political Research Quarterly, 62(3), 601–613. Sigelman, L. (2006). Report of the editor of the American Political Science Review, 2004–2005. PS: Political Science and Politics, 39, 171–173. Thornhill, R., & Palmer, C. T. (2000). A natural history of rape: Biological bases of sexual coercion. Cambridge, MA: The MIT Press. Warnke, G. (2007). After identity: Rethinking race, sex, and gender. Cambridge, UK: Cambridge University Press. Wilson, E. O. (1976). Sociobiology: The new synthesis. Cambridge, MA: The Belknap Press. Wilson, E. O. (1978). On human nature. Cambridge, MA: The Harvard University Press.
GENES, TWIN STUDIES, AND ANTISOCIAL BEHAVIOR Danielle Boisvert and Jamie Vaske INTRODUCTION While the field of criminology is rooted in sociological tradition, biosocial criminology has emerged as a promising perspective in studying antisocial behaviors. This perspective encompasses the research from other scientific disciplines, namely behavioral genetics and molecular genetics. At its core, biosocial criminology views criminal behavior as a function of both the social environment as well as biological/genetic factors. This chapter will provide a description of the prominent methodologies used in behavioral genetics and molecular genetics, a review of the empirical research, an overview of some of the statistical and methodological issues, as well as a discussion on the potential avenues for future research.
BEHAVIORAL GENETICS For years researchers have relied on the classical twin design to examine whether genetic factors influence antisocial behaviors. This method provides a crude estimate of heritability by taking twice the difference between the correlation coefficients of monozygotic (MZ) and dizygotic (DZ) twins for a particular behavior/trait of interest. Fortunately, more statistically Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 159–183 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009009
159
160
DANIELLE BOISVERT AND JAMIE VASKE
sophisticated approaches to studying genetic influences on behavior have been developed in the field of behavioral genetics. Known as biometrical modeling, this statistical procedure partitions the variance of a phenotype into its genetic and environmental components. There are three sources of genetic effects (i.e., additive, dominant, epistatic) that influence behaviors. While additive genetic effects refer to the sum of the average effects of all individual alleles, dominant and epistatic genetic effects refer to interactions between alleles at the same and different loci, respectively. All three of these genetic effects operate to make individuals within a family either more or less similar to one another (Lemery & Goldsmith, 1999). In addition to genetic effects, two sources of environmental influences, known as shared and nonshared, also affect behaviors. First, shared environmental influences (e.g., SES, neighborhood conditions) are believed to exert the same effect on children living within the same home thereby promoting behavioral similarities between siblings. Second, nonshared environmental influences (e.g., different peers, different hobbies) encompass experiences and life events that are unique to each child within the home, thereby promoting behavioral differences between siblings (Neale & Cardon, 1992).
ACE/ADE MODELS Researchers must first determine whether additive and/or dominant genetic effects are operating on a behavior. This is accomplished by examining the intraclass correlations of MZ and DZ twins on a specific behavior. If the intraclass correlation of a behavior for MZ twins is approximately two times that of DZ twins, this indicates that there are additive genetic effects influencing that behavior and the researcher would run an ACE model (see Eq. (1)). On the other hand, if the intraclass correlation of a behavior for MZ twins is more than two times that of DZ twins, this demonstrates that dominant genetic effects are also influencing the behavior and the researcher would run an ADE model (see Eq. (2)). VarP1 ¼ a2 þ c2 þ e2
(1)
VarP1 ¼ a2 þ d 2 þ e2
(2)
or
Where VarP1 symbolizes variance in phenotype 1, a2 proportion of the variance attributed to additive genetics, c2 proportion of the variance
161
Genes, Twin Studies, and Antisocial Behavior
attributed to shared environment, d2 proportion of the variance attributed to dominant genetics, and e2 proportion of the variance attributed to nonshared environment and measurement error. In the social sciences, ACE models are most commonly used in the estimation of genetic and environmental effects on behaviors. The structural equation modeling program, Mx, is equipped to estimate the degree to which genetic, shared, and nonshared environmental influences affect behaviors. As illustrated in Fig. 1, the correlation between the latent factors for additive genetics (A) is 1 for MZ twins because they share 100% of their DNA in common and 0.50 for DZ twins because they have approximately 50% of their DNA in common. The correlation between latent factors representing the shared environment (C) is 1 for both MZ and DZ twins because it is assumed that twin pairs will have 100% of their shared environment in common, regardless of their zygosity. The nonshared environment (E) is assumed to be unique to each individual and is therefore left uncorrelated between MZ and DZ twin pairs (Neale & Cardon, 1992). It is important to note that estimates of nonshared environment also include measurement error (Plomin, DeFries, McClearn, & Rutter, 1997). ACE models are first fitted to the data to create an overall saturated model. From there, two submodels (i.e., AE and CE) are created to compare against the saturated model in order to test the significance of both genetic and shared environmental factors. This is accomplished by individually setting the a and c parameters to zero and assessing the goodness-of-fit of the submodels by examining both the difference in chi-square statistic and the Akaike’s Information Criterion (AIC). The best fitting model is the one that has the largest negative AIC value and a nonsignificant p-value for the
1 or .5 1 1
1
A
1
C a
c
T1 Fig. 1.
1
E e
1
E
1
C e
c
A a
T2
Path Diagram of Univariate ACE Model.
162
DANIELLE BOISVERT AND JAMIE VASKE
difference in chi-square test. For example, if the AE submodel is considered the best fitting model, this implies that setting the c parameter to zero does not significantly decrease the fit of the data and that the effects of shared environmental factors on that behavior are negligible. The results from these analyses provide an overall estimate of the genetic and environmental effects on a behavior. It is important to note that these estimates will not provide any information on the specific environmental factors that are influencing behavior, such as parenting, neighborhood, education, etc. Nor will these estimates pinpoint the specific genes that are operating to affect behavior, such as 5-HTTLPR, DAT1, DRD2, DRD4, and monoamine oxidase A (MAOA) (which will be discussed in the Molecular Genetics section).
Review of Empirical Research Hundreds of twin studies have been conducted to estimate the degree to which genetic and environmental factors influence antisocial and delinquent behaviors (Moffitt, 2005). As mentioned, most studies conducted prior to the 1980s relied on crude methods of calculating heritability, such as using intraclass correlations. Recent twin studies, however, have relied on more sophisticated quantitative model-fitting approaches, such as ACE models. Overall, these studies provide consistent evidence that genetic factors influence variation in several forms of maladaptive behaviors, such as externalizing behaviors, aggression, conduct problems/disorders, antisocial behaviors, as well as delinquent and criminal behaviors. A review of this research is presented below. Externalizing Behaviors Externalizing behaviors are those that are directed outward, often toward another individual (e.g., disobedience, impulsive, aggressive). Many twin studies have assessed the genetic and environmental influences on externalizing behaviors. For example, Silberg et al. (1994) examined twin pairs from the Medical College of Virginia Adolescent Behavioral Development Twin Project. Their results revealed that for young boys between the ages of 8 and 11, 38%, 46%, and 16% of the variance in externalizing behaviors was accounted for by genetic, shared, and nonshared environmental factors, respectively. For young girls, on the other hand, 13%, 62%, and 25% of the variance in externalizing behaviors was accounted for by genetic, shared, and nonshared environmental factors,
Genes, Twin Studies, and Antisocial Behavior
163
respectively. These results suggest that genetic effects on externalizing behaviors are stronger in young boys compared to young girls. When comparing younger children to older youth, Schmitz, Fulker, and Mrazek (1995) found that genetic effects accounted for 34% of the variance in externalizing behaviors in children aged 2–3 while shared and nonshared environmental factors accounted for 32% and 34% of the variance, respectively. For youth between the ages of 4 and 18 years, 57% of the variance in externalizing behaviors was due to genetic factors, while shared and nonshared environmental factors accounted for the remaining 22% and 21% of the variance, respectively. Similar results were reported by Gjone and Stevenson in 1997 using data from the Norwegian Medical Birth Register. Specifically, they found that 38–47%, 47–50% and 7–12% of the variance in externalizing behaviors in younger kids between the ages of 5 and 9 was due to genetic, shared, and nonshared environmental factors, respectively. For older youth between the ages of 12 and 15 years, 57–65%, 23–33%, and 10–12% of the variance in externalizing behaviors was accounted for by genetic, shared, and nonshared environmental factors, respectively. Taken together, these studies suggest that genetic effects on externalizing behaviors are stronger in older youth compared to younger children.
Aggression Several studies have investigated the etiology of aggressive behaviors with a particular focus on the potential genetic factors that influence aggression. Using a sample of Canadian toddlers aged 19 months, Dionne, Tremblay, Boivin, Laplante, and Perusse (2003) reported that 58% of the variance in physical aggression, as measured by the child’s parents, was due to genetic factors while the remaining 42% was attributed to nonshared environmental factors. Using samples of Swedish and British youth, Eley, Lichtenstein, and Stevenson (1999) also assessed the genetic and environmental contributions to aggressive antisocial behaviors as measured by parental reports on the Child Behavior Checklist (CBCL). In the Swedish sample, they found that genetic factors accounted for 62–77% of the variance in aggressive behaviors while shared and nonshared environmental factors accounted for 4–12% and 18–27% of the variance, respectively. Similar results were reported with the British sample, where 50–76%, 10–18%, and 24–32% of the variance in aggressive behaviors was accounted for by genetic, shared, and nonshared environmental factors, respectively.
164
DANIELLE BOISVERT AND JAMIE VASKE
Using a sample of male twins in their mid-40s from the Vietnam Era Twin Register, Coccaro, Bergeman, Kavoussi, and Seroczynski (1997) assessed three measures of aggression from the Buss–Durkee Hostility Inventory (e.g., direct assault, indirect assault, verbal assault). They found that, depending on the measure of aggression, 28–47% of the variance was due to genetic factors while 53–72% of the variance was due to nonshared environmental factors. Also using a sample of adults, Finkel and McGue (1997) reported that genetic factors accounted for 35–39% of the variance in a measure of aggression from the Multidimensional Personality Questionnaire while the remaining 61–65% of the variance was attributed to nonshared environmental factors. These studies demonstrate that, although measured in a variety of ways, both genetic and environmental factors influence aggressive behaviors.
Conduct Problems/Disorder Conduct problems/disorder is often characterized as a pattern of behavior that violates the rights of others and/or violates society’s norms (e.g., lying, aggression, disobedience, truancy). Several studies have assessed the genetic and environmental influences on conduct problems/disorder in childhood and adolescence. For example, Burt, Krueger, McGue, and Iacono (2001) examined 1,506 twin pairs aged 11 from the Minnesota Twin Family Study. Their best fitting model revealed that 52%, 14%, and 34% of the variance in conduct disorder was attributed to genetic, shared, and nonshared environmental influences, respectively. Similar results were reported by Slutske et al. (1997) in their analyses of 2,682 twin pairs from the Australian Twin Registry. Participants self-reported their level of conduct problems, prior to the age of 15, using items derived from the DSM-III-R. Their results revealed that 71% of the variance in conduct problems was due to genetic factors with the remaining 29% being attributed to nonshared environmental factors. Researchers have also examined the genetic and environmental influences on conduct problems in adolescents. Young, Stallings, Corley, Krauter, and Hewitt (2000) recruited 344 pairs of twins aged 12–18 from the Colorado Longitudinal Twin Sample and the Colorado Twin Registry. Their analyses revealed that 34% and 66% of the variance in conduct disorder, measured using symptoms from the DSM-IV, was due to genetic and nonshared environmental factors, respectively. Miles, Van Den Bree, and Pickens (2002) also examined adolescents using data from the Add Health study and found that 61% of the variance in conduct disorders was attributed to
Genes, Twin Studies, and Antisocial Behavior
165
genetic factors while the remaining 39% was attributed to nonshared environmental factors. Antisocial Behaviors Several studies have assessed the genetic and environmental contributions to antisocial behaviors. For instance, using data from the Environmental Risk Longitudinal Twin Study, Arseneault et al. (2003) examined 1,116 pairs of twins aged 5 years. Multiple sources of information (i.e., mother, teacher, examiner-observer, and self-report) were used to assess the child’s level of antisocial behaviors. The results from their study revealed that 82% and 18% of the variance in a measure of antisocial behaviors that was agreed upon by all informants was due to genetic and nonshared environmental factors, respectively. When examining the results of all informants separately, genetic factors accounted for 42–76% of the variance in antisocial behaviors while the remaining 24–58% was attributed to nonshared environmental factors, depending on the source of information. Two measures of antisocial behaviors, based on observational data from the Nonshared Environmental and Adolescent Development (NEAD) Project, were included in a study by O’Connor and his colleagues in 1995. Specifically, the examiners reported the level of antisocial behaviors exhibited by the youth toward his/her mother and toward his/her father by observing their interactions. Their results revealed that genetic, shared, and nonshared environmental factors accounted to 36%, 10%, and 54% of the variance in antisocial behaviors directed toward the child’s mother, respectively. Similar results were reported for the variance in antisocial behaviors directed toward the child’s father with 31%, 17%, and 52% of the variance being attributed to genetic, shared, and nonshared environmental factors, respectively. In a follow-up study, O’Connor et al.(1998) also used the NEAD data to create a composite score for a measure of antisocial behaviors using three sources of information (i.e., self-report, parental report, and observational). They found that genetics, shared, and nonshared environmental factors accounted for 61%, 14%, and 25% of the variance in antisocial behaviors, respectively. Delinquency and Criminality Studies have also assessed the genetic and environmental influences on direct measures of delinquency and criminality. For instance, using data from the Minnesota Twin Registry, Krueger, Hicks, and McGue (2001) examined the genetic and environmental contributions to self-reported criminal behaviors in a sample of adults aged 33 years (N ¼ 397 twin pairs).
166
DANIELLE BOISVERT AND JAMIE VASKE
Their results revealed that 52% of the variance in the measure of criminality was due to genetic factors while the remaining 48% was due to nonshared environmental factors. Taylor, McGue, and Iacono (2000a) also assessed the genetic and environmental influences on a measure of delinquency using data from the Minnesota Twin Family Study. Participants, aged 16–18 years, self-reported whether they had ever engaged in a variety of minor and serious delinquent acts. Their best fitting model revealed that 18%, 26%, and 56% of the variance in delinquency was attributed to genetic, shared, and nonshared environmental factors, respectively. Furthermore, also using data from the Minnesota Twin Family Study, Taylor et al. (2000b) then examined whether there were sex differences in the genetic and environmental effects on delinquency for males and females. They found that, for males, genetic factors accounted for 55% of the variance in delinquency and the remaining 45% was attributed to nonshared environmental factors. On the other hand, for females, 8%, 37%, and 55% of the variance in delinquency was attributed to genetic, shared, and nonshared environmental factors, respectively. Although it appears that genetic factors had a stronger effect in males, this difference was not statistically significant.
Statistical and Methodological Issues Using structural equation modeling on twin data relies on the assumption that there is no assortative mating and that there are no gene–environment interactions influencing the trait of interest. First, assortative mating refers to the nonrandom manner in which individuals choose mates. Essentially, individuals tend to select sexual partners who have similar characteristics as themselves on some physical, cognitive, and/or personality trait. Research has shown that assortative mating can occur for antisocial behaviors. In other words, antisocial individuals tend to mate with other antisocial individuals (Caspi & Herbener, 1990; Quinton, Pickles, Maughan, & Rutter, 1993). The effects of nonrandom mating lead to increased similarities between DZ twins relative to MZ twins. From a statistical standpoint, this would lead to an underestimation of the genetic effects and overestimation of the shared environmental effects (Neale & Cardon, 1992). However, Maes et al. (1998) argue that the overall levels of assortative mating are quite low and that the biases associated with them are generally very small. In any event, future studies should use an extended-kinship study design that incorporates information from both twins and parents in order to explicitly test for the effects of assortative mating (Neale & Cardon, 1992).
Genes, Twin Studies, and Antisocial Behavior
167
The second assumption underlying the use of SEM with twin data is that there are no gene–environment interactions occurring. Gene–environment interactions refer to situations in which an individual’s genes interact with his/her environment to influence behavior. In other words, if an individual has a predisposition for antisocial behavior, that behavior will only be expressed when coupled with a particular environment. As illustrated in Fig. 1, there were no correlations (i.e., double-headed arrows) between A and C or between A and E, which infers that there were no gene– environment interactions occurring. However, as will be discussed in the Molecular Genetics section of this chapter, research has shown that gene– environment interactions are involved in the prediction for antisocial and criminal behaviors in both juveniles and adults (Caspi et al., 2002; Foley et al., 2004).
Future Research There are a number of potential avenues for the continued study of delinquent/criminal behaviors using a behavioral genetic approach. First, the use of twins when estimating the degree to which genetic factors influence behaviors has been called into question. Specifically, critics of behavioral genetic research argue that it is the similarity in treatment rather than genetics that accounts for the increased behavioral similarities found in MZ twins relative to DZ twins (Ellis, 1982). From a statistical standpoint, this increased similarity in behaviors for MZ twins would then overestimate the genetic influences on behavior. Therefore, future studies should strive to incorporate individuals of varying degrees of genetic relatedness into their models, such as parents and siblings. These extended designs provide more reliable estimations of the genetic and environmental contributions to delinquent/criminal behaviors. Another avenue of future research involves the exploration of sex differences in the genetic and environmental effects on delinquent/criminal behaviors. Gender is a robust predictor of criminal involvement with males participating in more criminal activities than females. Sex-limitation models can examine whether the same genetic and environmental factors are operating in men and women and whether the magnitude of the genetic and environmental effects is the same across genders (Neale & Cardon, 1992). While studies have used sex-limitation models in the examination of various forms of antisocial behaviors, few studies have examined whether there are genetic sex differences in direct measures of delinquent/criminal behaviors
168
DANIELLE BOISVERT AND JAMIE VASKE
(see Boisvert, 2009; Eley et al., 1999; Taylor et al., 2000b). Therefore, future studies should consider sex-limitation models when estimating the genetic and environmental influences on delinquent/criminal behaviors. Furthermore, delinquent/criminal behaviors often co-occur with a variety of other antisocial behaviors, such as externalizing behaviors, aggression, and conduct problems/disorders. Studies have examined the genetic and environmental effects on the co-occurrence of a variety of maladaptive behaviors, such as aggressive and nonaggressive forms of antisocial behaviors (Button, Scourfield, Martin, & McGuffin, 2004) as well as conduct problems and marijuana/alcohol use (True et al., 1999). However, there remains a gap in the literature, and a need for further research, on the degree to which the co-occurrence of delinquent/criminal behaviors and other antisocial behaviors (e.g., aggression, conduct problems) is explained by the same genetic and environmental factors operating on both types of behaviors.
MOLECULAR GENETICS Behavioral genetic studies have revealed that genetic factors explain a moderate to large percentage of variance in antisocial behaviors. As mentioned, however, these studies cannot tell researchers which genetic factors are associated with antisocial/criminal behavior. Instead, researchers must turn to the field of molecular genetics to examine this question. Molecular genetics is the study of gene functioning and how individual differences in gene functioning influences one’s health, cognition, and behavior (see Lewis, 2001 for an overview of molecular genetics). In particular, researchers who have examined the link between genes and criminal behavior have predominantly focused on genetic polymorphisms related to neuropsychological systems, such as the serotonin system, the dopamine system, and enzymatic system. It is believed that these genetic polymorphisms influence neuropsychology and cognition (i.e., how or what a person thinks), which in turn influences a person’s behavior. Three different types of molecular genetics studies are often used: (1) gene-behavior studies in which genetic markers are directly associated with antisocial behavior; (2) gene gene studies in which the effects of one genetic risk factor on criminal behavior is conditioned or moderated by a second genetic risk factor; and (3) gene environment studies in which genetic markers are associated with delinquent or criminal behavior when they are paired with a certain environment. The following section will review
Polymorphism Name
13q14-q21
13q14-q21
17q11.1-17q12
Chromosome
G-A substitution
C-T substitution
44 bp insertion/ deletion polymorphism
Polymorphism
G allele, A allele
His-allele Tyr allele
Long allele, short allele
Alleles
Functional Significance
Information on Candidate Genes for Delinquent and Criminal Behavior.
5p15.3
Short allele associated with reduced transcription of 5HTT gene, which results in less serotonin transporter protein, fewer serotonin transporters, and subsequently less reuptake of serotonin Tyr allele associated with reduced signal transduction in 5HT2A receptors (i.e., movement of a message from outside a cell to inside the cell) Not known, but it is hypothesized that the polymorphism reduces the number of receptors by influencing transcription 9-repeat allele associated with reduced transcription of DAT1 gene, which results in less dopamine transporter protein, fewer dopamine transporters, and subsequently less reuptake of dopamine A1 allele reduces the density of dopamine receptors
Genes, Twin Studies, and Antisocial Behavior
2-repeat, 11repeat alleles 2-, 3-, 3.5-, 4-, 5-repeat alleles
Not known, but it is hypothesised so that the 7 þ -repeat alleles are associated with reduced transcription of the DRD4 gene, less efficient protein folding, and subsequently fewer dopamine D4 receptors 2-, 3-, and 5-repeat alleles associated with reduced transcription of the MAOA gene, which may result in less degradation of neurotransmitters in synaptic cleft
9-repeat allele, 10repeat allele A2 allele, A1 allele
11q23
11p15.5
Xp11.3-11.4
48 bp variable number tandem repeat polymorphism 30 bp variable number tandem repeat polymorphism
40 bp variable number tandem repeat polymorphism C-T substitution
Table 1.
Serotonin transporter gene, 5-HTTLPR Serotonin receptor gene, 5HT2A His452Tyr Serotonin receptor gene, 5HT2A 1438G/A Dopamine transporter gene, DAT1 3uUTR VNTR Dopamine receptor gene, DRD2 TaqIA RLFP Dopamine receptor gene, DRD4 48 bp VNTR Monoamine oxidase A gene, MAOA 30 bp VNTR
169
Table 1. (Continued )
L allele, U allele
Alleles
A-C substitution
Polymorphism
11p15.3-p14
Val allele, Met allele
Chromosome
22q11
G-A substitution
Functional Significance
Not established. Polymorphism may influence splicing or stability of mRNA. It is likely that the A218C polymorphism is not functional, but it is in linkage disequilibrium with the A779C functional polymorphism Val allele is associated with increased enzymatic activity in the prefrontal cortex, and subsequently lower levels of dopamine activity in the prefrontal cortex
170
Tryptophan hydroxylase gene, A218C
Polymorphism Name
Catechol-Omethyltransferase gene (COMT Val158Met)
The 40bp VNTR can be repeated 3–11 times. The 9R and 10R allele are the most commonly occurring alleles.
DANIELLE BOISVERT AND JAMIE VASKE
Genes, Twin Studies, and Antisocial Behavior
171
the empirical evidence linking specific genetic polymorphisms to delinquent and criminal behavior using these three types of studies (Table 1).
Review of Empirical Research Serotonin Transporter Polymorphism (5-HTTLPR) A number of gene-behavior studies have found that the serotonin transporter polymorphism, 5-HTTLPR, is significantly related to delinquent and criminal behavior. For instance, Liao, Hong, Shih, and Tsai (2004) found that violent offenders were more likely to carry one or more copies of the short 5-HTTLPR allele compared to healthy controls in a sample of Chinese men. Also, a study of Caucasian incarcerated males revealed that the prevalence of the short allele of 5-HTTLPR was significantly higher among violent inmates compared to nonviolent inmates (Retz, Retz€ Junginger, Supprian, Thome, & Rosler, 2004). Using data from the National Longitudinal Study of Adolescent Health (Add Health), Beaver, Wright, DeLisi, and Vaughn (2008a) revealed that individuals who carried the short allele of 5-HTTLPR were significantly less likely to desist from crime during adolescence and adulthood while Vaughn, DeLisi, Beaver, and Wright (2009a) reported a positive relationship between 5-HTTLPR and chronic offending in the African American subsample. Research has also found that 5-HTTLPR interacts with other genetic markers and the environment to influence delinquent and criminal behavior. For example, Rae (2006) found that the expected rate of arrest was significantly higher among individuals who carried a short allele of 5-HTTLPR and who were homozygous for the 10-repeat allele at the DAT1 gene compared to individuals who carried a short 5-HTTLPR allele and who were not homozygous for the 10-repeat allele at DAT1. In addition, a study of 184 adult males referred for a forensic evaluation revealed that 5-HTTLPR interacted with childhood adversity to influence violent criminal behavior in adulthood (Reif et al., 2007). Vaske et al.’s (2009) analysis of the Add Health data revealed that marijuana use leads to higher levels of property offending among African American females when respondents carried one or more copies of the short 5-HTTLPR allele. Serotonin Receptor Polymorphisms (5HT2A His452Tyr and 5HT2A 1438G/A) Researchers have recently begun to examine the role of serotonin receptor polymorphisms in the etiology of delinquent and criminal behavior.
172
DANIELLE BOISVERT AND JAMIE VASKE
Burt and Mikolajewski’s (2008) analysis of 211 undergraduate males revealed that males who carried two copies of the His-allele in 5HT2A His452Tyr reported higher levels of nonaggressive delinquent behaviors than males who were not homozygous for the His-allele. However, their analyses also revealed that 5HT2A His452Tyr was not significantly related to aggressive forms of delinquent behavior. A follow-up study by Burt (2009) found that males who carried one or more copies of the G allele reported higher levels of delinquent and antisocial behavior, and these males were subsequently rated as more popular by their peers. On the other hand, Berggard et al. (2003) found that the prevalence of the A allele of the 5HT2A 1438G/A polymorphism was significantly higher among offenders than nonoffenders in a Swedish sample.
Dopamine Transporter Polymorphism (DAT1 3uUTR VNTR) Several studies have examined the association between the dopamine transporter polymorphism, DAT1, and criminal behavior. For instance, Burt and Mikolajewski (2008) found that undergraduate males who engaged in nonaggressive forms of delinquent behavior were more likely to carry two copies of the 10-repeat allele of DAT1. Studies using data from the Add Health study have found that the prevalence of the 10-repeat allele of DAT1 is significantly higher among individuals who engaged in higher levels of criminal behavior (Beaver, Wright, & Walsh, 2008b), serious delinquency (Guo, Roettger, & Shih, 2007), violent delinquency (Guo, Roettger, & Cai, 2008a), as well as individuals who have more contact with the police (Vaughn et al., 2009a) and who are less likely to desist from criminal behavior (Beaver et al., 2008a). Other studies, however, have reported either no relationship between the DAT1 polymorphism and criminal behavior (Reif et al., 2007; Caspi et al., 2008), or that the 9-repeat allele of DAT1 is associated with criminal behavior (Gerra et al., 2005). Thus, it is not yet clear whether the 10- or 9-repeat allele of DAT1 connotes greater risk for delinquent and criminal behavior. Studies have also found that DAT1 interacts with other genetic polymorphisms and negative environments to influence criminal behavior. As previously discussed, Rae (2006) found that 5-HTTLPR and DAT1 interacted to influence the total number of arrests in an at-risk sample of young African American respondents. Furthermore, Vaughn et al. (2009a) found that having few delinquent peers interacted with the 10-repeat allele of DAT1 to predict serious violent criminal behavior for Caucasian respondents.
Genes, Twin Studies, and Antisocial Behavior
173
Dopamine Receptor Polymorphism (DRD2 TaqIA RFLP) Researchers have also investigated the relationship between the dopamine receptor polymorphism, DRD2, and criminal behavior. For example, Guo et al. (2007) found that males who had the A1/A2 genotype at DRD2 reported significantly higher levels of serious delinquency and violent delinquency than males who carried two copies of the A1 allele or two copies of the A2 allele at DRD2. Vaske (2009) also found that DRD2 was associated with violent criminal behavior during adulthood for males but not for females. Similarly, Boutwell and Beaver (2008) reported that males with the A1 allele at DRD2 were less likely to abstain from criminal behavior, but that DRD2 was not related to criminal behavior for females. In addition to these gene-behavior studies, research has also shown that DRD2 interacts with other genetic markers and criminogenic environments to influence delinquent and criminal behavior. For example, Beaver et al. (2007) found that DRD2 interacted with DRD4 to predict delinquent and criminal behavior among males in the Add Health study. Furthermore, Guo et al.’s (2008a) analysis of 1,100 males from the Add Health study revealed that the expected count of serious and violent delinquency was highest among respondents who had the A1/A2 genotype at DRD2 and who did not have regular meals with a parent, who did not grow up with two biological parents in the home, who reported that social services had investigated their parents, and who had more deviant peers. Overall, their findings suggest that delinquency was highest among individuals who carried the DRD2 risk allele and who had weak parental bonds. On the contrary, DeLisi, Beaver, Wright, and Vaughn (2008) found that DRD2 was significantly associated with delinquency (e.g., age of first contact with police and age of first arrest) for those who reported strong parental bonds. The differences in results may be a consequence of the analytical samples used in each study (i.e., Guo et al. (2008a) used data from all males in the Add Health, while DeLisi et al. (2008) used data only from respondents who reported contact with police) as well as analytical techniques (i.e., Guo et al. used a hierarchical model to account for the nesting of subjects within level 2 units, while DeLisi et al. used Huber/White corrections). Dopamine Receptor Polymorphism (DRD4 48 bp) Researchers have also examined the relationship between another dopamine receptor polymorphisms, DRD4, and delinquent/criminal behavior. However, the current research is mixed and inconclusive. For example, Boutwell and Beaver’s (2008) analysis of males revealed that those who did not abstain from delinquent behavior were more likely to carry a 7-repeat allele
174
DANIELLE BOISVERT AND JAMIE VASKE
at DRD4 than males who did abstain from delinquent behavior. Furthermore, Dmitrieva, Chen, Greenberger, Ogunseitan, and Ding (2010) found that males with the 7-repeat allele at DRD4 reported higher levels of delinquency than males with any of the lower repeat alleles at DRD4. This effect, however, was not found for females. The authors also found that parental monitoring and exposure to violence reduced the interaction between gender and DRD4 on delinquency to nonsignificance; thus, DRD4 was more likely to lead to delinquent behavior for boys (compared to girls) because boys experienced lower levels of parental monitoring and greater exposure to violence. Other studies, however, have found that DRD4 was not related to delinquent or criminal behavior (Vaske, 2009). For instance, Caspi et al. (2008) found that DRD4 was not significantly associated with adult criminal conviction among children with attention deficit hyperactivity disorder. Further, Rae (2006) revealed that DRD4 was negatively associated with violent criminal behavior in adulthood. That is, individuals who carried a 7 W repeat allele were less likely to engage in violent criminal behavior, while individuals with a lower repeat allele were more likely to violently offend. At this point in time, the extent to which DRD4 is related to delinquent and criminal behavior is not yet clear. Monoamine Oxidase a Polymorphism (MAOA 30 bp VNTR) Several studies have shown that polymorphisms in the MAOA gene are related to delinquent and criminal behavior. Reif et al.’s (2007) analysis of clinically referred males revealed that males who engaged in violent criminal behavior were more likely to carry an MAOA low-activity allele (i.e., 2- or 3-repeat allele) compared to males who did not engage in violent criminal behavior. Other studies of clinical populations have also revealed that the MAOA low-activity allele is more prevalent among individuals who engage in antisocial behavior than nonoffenders and healthy controls (Koller, Bondy, Preuss, Bottlender, & Soyka, 2003; Samochowiec et al., 1999). Using samples from the general population, studies have found that individuals who carried the MAOA low-activity alleles reported higher levels of serious and violent delinquency in adolescence and adulthood (Guo et al., 2008a, 2008b). It is important to note that other studies have failed to find a significant relationship between the MAOA polymorphism and criminal behavior (Lu, Lin, Lee, Ko, & Shih, 2003; Vaske, 2009). Several gene environment studies have revealed that the MAOA polymorphism moderates the effect of criminogenic environments on delinquent and criminal behavior. For instance, Caspi et al.’s (2002) hallmark study of males in the Dunedin Multidisciplinary sample found that
Genes, Twin Studies, and Antisocial Behavior
175
MAOA interacted with childhood maltreatment to predict delinquent and criminal behavior. The authors found that maltreated males who carried an MAOA low-activity allele were more likely to engage in delinquent behavior, have an adult conviction for a violent offense, self-report higher levels of violent behavior, and exhibit symptoms of antisocial personality disorder. Subsequent studies have also found a significant interaction between MAOA and childhood maltreatment in predicting antisocial behavior (Edwards et al., 2009; Widom & Brzustowicz, 2006), yet others have failed to replicate these findings (Haberstick et al., 2005; Young et al., 2006; Vaske, 2009).
Tryptophan Hydroxylase Polymorphism (TPH A218C) To our knowledge, there has been only one study that has investigated the link between the TPH A218C polymorphism and criminal behavior. Specifically, Burt and Mikolajewski’s (2008) analysis of 211 undergraduate males revealed that males who carried the L allele did not report higher levels of criminal behavior than males with the U allele. At this point in time, further studies are needed to determine whether the TPH A218C polymorphism is related to delinquent and criminal behavior. Catechol-O-Methyltransferase Gene (COMT Val158Met) Recent research has also examined the relationship between the COMT Val158Met polymorphism and criminal behavior using clinical populations. For example, Caspi et al. (2008) found that individuals who were diagnosed with ADHD during childhood and who carried two copies of the Val allele were more likely to be convicted of criminal offenses during adulthood than ADHD individuals who carried the Met allele. Other studies, however, have found that it is the Met allele that is associated with criminal behavior. Specifically, Kotler et al.’s (1999) analysis of incarcerated schizophrenic patients found that the low-activity Met allele was more prevalent among homicidal schizophrenic patients than nonviolent schizophrenic patients and healthy controls. Strous, Bark, Parsia, Volavka, and Lachman (1997) also found that the Met allele was more prevalent among schizophrenic patients who engaged in violent criminal behavior and substance abuse than schizophrenic patients who did not have a history of violent offending. While these studies show that COMT is associated with criminal behavior in clinical populations, it remains unclear whether COMT is associated with an increased risk of offending in the general population.
176
DANIELLE BOISVERT AND JAMIE VASKE
Statistical and Methodological Issues While the majority of these studies show that various genetic polymorphisms are significantly related to criminal behavior, there are a number of limitations that need to be discussed. First, it is likely that many of the individual genetic polymorphisms only account for a small percentage of variation in criminal behavior, which may explain why some studies fail to find a significant relationship between a genetic polymorphism and criminal behavior. Instead, it is much more likely that criminal behavior is a polygenic phenotype in that hundreds or thousands of genetic polymorphisms combine in additive and multiplicative ways to influence antisocial behaviors. Second, many of the gene environment interaction studies have a small number of cases falling into the interaction categories. This may cause studies to lack sufficient power to detect small or medium effect sizes and the results may be unstable. It is recommended that researchers estimate how many cases are needed to have acceptable power (i.e., 80% power). Third, researchers have called for better, more precise measures of environmental factors. For instance, future research should: (a) differentiate between long-term exposure to a risk factor versus short-term exposure; (b) consider age appropriate environmental risk factors; (c) avoid using retrospective measures of environmental risk factors, as these measures may be unreliable and contain a substantial amount of measurement error; and (d) avoid using composite or aggregate measures of environmental risk as they may mask the effects of individually relevant risk factors (Moffitt, Caspi, & Rutter, 2006). Finally, the last two methodological issues are related to linkage disequilibrium and genotypes that fail to meet the Hardy–Weinberg equilibrium (HWE). Linkage disequilibrium occurs when two or more genetic polymorphisms are simultaneously transmitted or inherited together, which subsequently causes a nonrandom association between the polymorphisms. For instance, research suggests that the TPH A218C polymorphism is in linkage disequilibrium with the TPH A779C polymorphism (Nielsen, Jenkins, Stefanisko, Jefferson, & Goldman, 1997). Linkage disequilibrium is important because if a nonfunctional polymorphism (i.e., the polymorphism does not directly specify or code for a protein) is in linkage with a functional polymorphism, then any association between a nonfunctional polymorphism and a phenotype, such as criminal behavior, may actually reflect the effect of the functional polymorphism; thus, the nonfunctional polymorphism is acting as a proxy for the functional polymorphism. This is important if researchers are interested in pinpointing
Genes, Twin Studies, and Antisocial Behavior
177
functional polymorphisms that contribute to a disorder, disease, or behavior. Linkage disequilibrium is also relevant because it will cause genotypes to deviate from HWE. HWE occurs when the frequency of alleles stay constant from one generation to the next. From a theoretical standpoint, HWE is important because it allows researchers to track the evolution of genes, and it allows researchers to estimate the probability of offspring(s) having a disorder or disease. From a statistical standpoint, HWE is important because deviations from HWE may indicate that a sample is not genetically representative of the population.
Future Direction There are a number of future directions for the investigation of molecular genetics and delinquent/criminal behavior. First, researchers are beginning to integrate powerful statistical techniques, such as structural equation models and multilevel models, with molecular genetic and criminological data when investigating the relationship between genetic polymorphisms and criminal behavior (Burt, 2009; Vaughn, Beaver, & DeLisi, 2009b). It is likely that researchers will continue this trend and expand their statistical analyses to other techniques, such as latent trajectory modeling. Second, it is likely that the field of criminology will begin to focus on the epigenetic effects of criminogenic environments on genetic expression and behavior. An epigenetic effect occurs when the environment either speeds up or slows down the expression of a gene, and this change in genetic expression subsequently influences behavior. There is a growing body of evidence from the animal literature which shows that exposure to abusive/stressful environments early in life influences the expression of genes related to the hypothalamic–pituitary–adrenal axis, which influences behaviors, such as antisocial behaviors (Andersen & Teicher, 2008). Third, it is likely that criminologists will continue to investigate the interaction between genetic polymorphisms and other biological processes, such as sex hormones. € Sjoberg et al. (2008) recently found that MAOA interacted with testosterone to influence aggression among males while Beaver, DeLisi, Vaughn, and Wright (2010) found that MAOA interacted with neuropsychological deficits to predict delinquency among Caucasian males. These studies suggest that genetic polymorphisms may moderate the effects of other biologically based processes on criminal behavior. Finally, another avenue for investigation is to examine whether the effectiveness of correctional treatment programs differ based on an
178
DANIELLE BOISVERT AND JAMIE VASKE
individual’s genotype. To our knowledge only one study has examined whether genetic polymorphisms moderate the effectiveness of an intervention program. Specifically, Bakersmans-Kranenburg, VanIjzendoorn, Mesman, Alink, and Juffer (2008) found that a parenting skills intervention program successfully reduced cortisol levels among preschool children who carried the 7-repeat allele of the DRD4 polymorphism, but not among children who did not carry the 7-repeat allele. The authors suggested that lowering cortisol levels among the preschool children may have led to lower levels of externalizing behavior, but this hypothesis was not formally tested.
CONCLUSION In sum, biosocial criminology is an emerging perspective in the field of criminology and has the potential to seriously change the way we think about and study criminal behavior. Following a multidisciplinary approach, biosocial criminology incorporates knowledge and research from other scientific disciplines such as psychiatry, neurology, evolutionary psychology, and most importantly behavioral genetics and molecular genetics (Wright & Boisvert, 2009). Research using behavioral genetic methodology has clearly demonstrated that both genetic and environmental factors influence antisocial/criminal behaviors. On average, it appears that approximately 42%, 16%, and 42% of the variance in antisocial behavior is attributed to genetic (i.e., additive and nonadditive), shared, and nonshared environmental factors, respectively (see Rhee & Waldman, 2002). Furthermore, researchers using molecular genetic methodology have pinpointed specific genes to consider in the study of antisocial behavior. These genetic polymorphisms include a serotonin transporter (5-HTTLPR), two serotonin receptors (5HT2A His452Tyr and 5HT2A 1438G/A), a dopamine transporter (DAT1 3uUTR VNTR), two dopamine receptors (DRD2 TaqIA RFLP and DRD4 48 bp VNTR), monoamine oxidase A (MAOA 30 bp VNTR), tryptophan hydroxylase (TPH A218C), and catechol-O-methyltransferase (COMT Val158Met). The research presented here reinforces the notion that biosocial criminology is advancing the field of criminology by incorporating (rather than ignoring) genetic risk factors in the study of antisocial and criminal behaviors. Although several advancements have been made, there are still many new and exciting opportunities for continued research in this area.
Genes, Twin Studies, and Antisocial Behavior
179
REFERENCES Andersen, S. L., & Teicher, M. H. (2008). Desperately driven and no brakes: Developmental stress exposure and subsequent risk for substance abuse. Neuroscience and Biobehavioral Reviews, 33, 516–524. Arseneault, L., Moffitt, T. E., Caspi, A., Taylor, A., Rijsdijk, F. V., Jaffee, S. R., et al. (2003). Strong genetic effects on cross-situational antisocial behavior among 5-year-old children according to mothers, teachers, examiner-observers, and twins’ self-reports. Journal of Child Psychology and Psychiatry and Allied Discipline, 44, 832–848. Bakersmans-Kranenburg, M. J., VanIjzendoorn, M. H., Mesman, J., Alink, L. R. A., & Juffer, F. (2008). Effects of an attachment-based intervention on daily cortisol moderated by dopamine receptor D4: A randomized control trial on 1- to 3-year-olds screened for externalizing behavior. Development and Psychopathology, 20, 805–820. Beaver, K. M., DeLisi, M., Vaughn, M. G., & Wright, J. P. (2010). The intersection of genes and neuropsychological deficits in the prediction of adolescent delinquency and low selfcontrol. International Journal of Offender Therapy and Comparative Criminology, 54, 22–42. Beaver, K. M., Wright, J. P., DeLisi, M., & Vaughn, M. G. (2008a). Desistance from delinquency: The marriage effect revisited and extended. Social Science Research, 37, 736–752. Beaver, K. M., Wright, J. P., & Walsh, A. (2008b). A gene-based evolutionary explanation for the association between criminal involvement and number of sex partners. Biodemography and Social Biology, 54, 47–55. Beaver, K. M., Wright, J. P., DeLisi, M., Walsh, A., Vaughn, M. G., Boisvert, D., & Vaske, J. (2007). A gene gene interaction between DRD2 and DRD4 is associated with conduct disorder and antisocial behavior in males. Behavioral and Brain Functions, 3, 30. Berggard, C., Damberg, M., Longato-Stadler, E., Hallman, J., Oreland, L., & Garpenstrand, H. (2003). The serotonin 2A-1438 G/A receptor polymorphism in a group of Swedish male criminals. Neuroscience Letters, 347, 196–198. Boisvert, D. (2009). Rethinking Gottfredson and Hirschi’s general theory of crime: A behavioral genetic approach. Unpublished thesis, University of Cincinnati, Cincinnati, OH. Boutwell, B. B., & Beaver, K. M. (2008). A biosocial explanation of delinquency abstention. Criminal Behaviour and Mental Health, 18, 59–74. Burt, S. A. (2009). A mechanistic explanation of popularity: Genes, rule breaking, and evocative gene–environment correlations. Journal of Personality and Social Psychology, 96, 783–794. Burt, S. A., Krueger, R. F., McGue, M., & Iacono, W. G. (2001). Sources of covariation among attention-deficit/hyperactivity disorder, oppositional defiant disorder, and conduct disorder: The importance of shared environment. Journal of Abnormal Psychology, 110, 516–525. Burt, S. A., & Mikolajewski, A. J. (2008). Preliminary evidence that specific candidate genes are associated with adolescent-onset antisocial behavior. Aggressive Behavior, 34, 437–445. Button, T. M., Scourfield, J., Martin, N., & McGuffin, P. (2004). Do aggressive and nonaggressive antisocial behavior in adolescents result from the same genetic and environmental effects?. American Journal of Medical Genetics, 129B, 59–63.
180
DANIELLE BOISVERT AND JAMIE VASKE
Caspi, A., & Herbener, E. S. (1990). Continuity and change: Assortative marriage and the consistency of personality in adulthood. Journal of Personality and Social Psychology, 58, 250–258. Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., et al. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297, 851–854. Caspi, A., Langley, K., Milne, B., Moffitt, T. E., O’Donovan, M., Owen, M. J., et al. (2008). A replicated molecular genetic basis for subtyping antisocial behavior in children with attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 65, 203–210. Coccaro, E. F., Bergeman, C. S., Kavoussi, R. J., & Seroczynski, A. D. (1997). Heritability of aggression and irritability: A twin study of the Buss–Durkee aggression scales in adult male subjects. Biological Psychiatry, 41, 273–284. DeLisi, M., Beaver, K. M., Wright, J. P., & Vaughn, M. G. (2008). The etiology of criminal onset: The enduring salience of nature and nurture. Journal of Criminal Justice, 36, 217–223. Dionne, G., Tremblay, R., Boivin, M., Laplante, D., & Perusse, D. (2003). Physical aggression and expressive vocabulary in 19-month-old twins. Developmental Psychology, 39, 261–273. Dmitrieva, J., Chen, C., Greenberger, E., Ogunseitan, O., & Ding, Y. C. (2010). Gender specific expression of the DRD4 gene on adolescent delinquency, anger, and thrill seeking. Social Cognitive and Affective Neuroscience (epub). Edwards, A. C., Dodge, K. A., Latendresse, S. J., Lansford, J. E., Bates, J. E., Pettit, G. S., Budde, J. P., Goate, A. M., & Dick, D. M. (2009). MAO-uVNTR and early physical discipline interact to influence delinquent behavior. Journal of Child Psychology and Psychiatry (epub). Eley, T. C., Lichtenstein, P., & Stevenson, J. (1999). Sex differences in the etiology of aggressive and nonaggressive antisocial behavior: Results from two twin studies. Child Development, 70, 155–168. Ellis, L. (1982). Genetics and criminal behavior. Criminology, 20, 43–66. Finkel, D., & McGue, M. (1997). Sex differences and nonadditivity in heritability of the multidimensional personality questionnaire scales. Journal of Personality and Social Psychology, 72, 929–938. Foley, D. L., Eaves, L. J., Wormley, B., Silberg, J. L., Maes, H. H., Kuhn, J., & Riley, B. (2004). Childhood adversity, monoamine oxidase A genotype, and risk for conduct disorder. Archives of General Psychiatry, 21, 738–744. Gerra, G., Garofano, L., Pellegrini, C., Bosari, S., Zaimovic, A., Moi, G., Avanzini, P., Talarico, E., Gardini, F., & Donnini, C. (2005). Allelic association of a dopamine transporter gene polymorphism with antisocial behavior in heroin-dependent patients. Addiction Biology, 10, 275–281. Gjone, H., & Stevenson, J. (1997). A longitudinal twin study of temperament and behavior problems: Common genetic or environmental influences? Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1448–1456. Guo, G., Roettger, M. E., & Shih, J. C. (2007). Contributions of the DAT1 and DRD2 genes to serious and violent delinquency among adolescents and young adults. Human Genetics, 121, 125–136. Guo, G., Roettger, M. E., & Cai, T. (2008a). The integration of genetic propensities into social control models of delinquency and violence among male youths. American Sociological Review, 73, 543–568.
Genes, Twin Studies, and Antisocial Behavior
181
Guo, G., Ou, X., Roettger, M. E., & Shih, J. C. (2008b). The VNTR 2 repeat in MAOA and delinquent behavior in adolescence and young adulthood: Associations and MAOA promoter activity. European Journal of Human Genetics, 16, 626–634. Haberstick, B. C., Lessem, J. M., Hopfer, C. J., Smolen, A., Ehringer, M. A., Timberlake, D., & Hewitt, J. K. (2005). Monoamine oxidase A (MAOA) and antisocial behaviors in the presence of childhood and adolescent maltreatment. American Journal of Medical Genetics Part B, 135(B), 59–64. Koller, G., Bondy, B., Preuss, U. W., Bottlender, M., & Soyka, M. (2003). No association between a polymorphism in the promoter region of the MAOA gene with antisocial personality traits in alcoholics. Alcohol and Alcoholism, 38, 31–34. Kotler, M., Barak, P., Cohen, H., Averbuch, I. E., Grinshpoon, A., Gritsenko, I., & Ebstein, R. P. (1999). Homicidal behavior in schizophrenia associated with a genetic polymorphism determining low catechol O-methyltransferase (COMT) activity. American Journal of Medical Genetics, 88, 628–633. Krueger, R. F., Hicks, B. M., & McGue, M. (2001). Altruism and antisocial behavior: Independent tendencies, unique personality correlates, distinct etiologies. Psychological Science, 12, 397–402. Lemery, K. S., & Goldsmith, H. H. (1999). Genetically informative designs for the study of behavioural development. International Journal of Behavioral Development, 23, 293–317. Lewis, R. (2001). Human genetics: Concepts and applications. New York: McGraw Hill. Liao, D., Hong, C., Shih, H., & Tsai, S. (2004). Possible association between serotonin transporter promoter region polymorphism and extremely violent crime in Chinese males. Neuropsychobiology, 50, 284–287. Lu, R., Lin, W., Lee, J., Ko, H., & Shih, J. C. (2003). Neither antisocial personality disorder nor antisocial alcoholism is associated with the MAO-A gene in Han Chinese males. Alcoholism: Clinical and Experimental Research, 27, 889–893. Maes, H. H. M., Neale, M. C., Kendler, K., Hewitt, J. K., Silberg, J. L., Foley, D. L., Meyer, J. M., Rutter, M., Simonoff, E., Pickles, A., & Eaves, L. J. (1998). Assortative mating for major psychiatric diagnosis in two population-based samples. Psychological Medicine, 28, 389–1401. Miles, D. R., Van Den Bree, V. B., & Pickens, R. W. (2002). Sex differences in shared genetic and environmental influences between conduct disorder symptoms and marijuana use in adolescents. American Journal of Medical Genetics, 114, 159–168. Moffitt, T. E. (2005). Genetic and environmental influences on antisocial behaviors: Evidence from behavioral-genetic research. Advances in Genetics, 55, 41–104. Moffitt, T. E., Caspi, A., & Rutter, M. (2006). Measured gene-environment interactions in psychopathology: Concepts, research strategies, and implications for research, intervention, and public understanding of genetics. Perspectives on Psychological Science, 1, 5–27. Neale, M. C., & Cardon, L. R. (1992). Methodology for genetic studies of twins and families. Dordrecht: Kluwer Academic. Nielsen, D. A., Jenkins, G. L., Stefanisko, K. M., Jefferson, K. K., & Goldman, D. (1997). Sequence, splice site and population frequency distribution analyses of the polymorphic human tryptophan hydroxylase intron 7. Molecular Brain Research, 45, 145–148. O’Connor, T. G., Hetherington, E. M., Reiss, D., & Plomin, R. (1995). A twin-sibling study of observed parent-adolescent relations. Child Development, 66, 812–829.
182
DANIELLE BOISVERT AND JAMIE VASKE
O’Connor, T. G., Neiderhiser, J. M., Reiss, D., Hetherington, E. M., & Plomin, R. (1998). Genetic contributions to continuity, change, and co-occurrence of antisocial and depressive symptoms in adolescence. Journal of Child Psychology and Psychiatry, 39, 323–336. Plomin, R., DeFries, J. C., McClearn, G. E., & Rutter, M. (1997). Behavioral genetics. New York: W. H. Freeman. Quinton, D., Pickles, A., Maughan, B., & Rutter, M. (1993). Partners, peers, and pathways: Assortative pairing and continuities in conduct disorder. Development and Psychopathology, 5, 763–783. Rae, M. N. (2006). Genetic and environmental influences on criminal behavior. Unpublished thesis, University of Cincinnati, Cincinnati, OH. € Reif, A., Rosler, M., Freitag, C. M., Schneider, M., Eujen, A., Kissling, C., Wenzler, D., Jacob, C. P., Junginger, P. R., Thome, J., Lesch, K. P., & Retz, W. (2007). Nature and nurture predispose to violent behavior: Serotonergic genes and adverse childhood environment. Neuropsychopharmacology, 32, 2375–2383. € Retz, W., Retz-Junginger, P., Supprian, T., Thome, J. M., & Rosler, M. (2004). Association of serotonin transporter promoter gene polymorphism with violence: Relation with personality disorders, impulsivity, and childhood ADHD psychopathology. Behavioral Sciences and the Law, 22, 415–425. Rhee, S. H., & Waldman, I. D. (2002). Genetic and environmental influences on antisocial behavior: A meta-analysis of twin and adoption studies. Psychological Bulletin, 128, 490–529. Samochowiec, J., Lesch, K. P., Rottmann, M., Smolka, M., Sagailo, Y. V., Okladnova, O., Rommelspacherc, H., Wintererb, G., Schmidtb, L. G., & Sanderbd, T. (1999). Association of a regulatory polymorphism in the promoter region of the monoamine oxidase A gene with antisocial alcoholism. Psychiatry Research, 86, 67–72. Schmitz, S., Fulker, D. W., & Mrazek, D. A. (1995). Problem behavior in early and middle childhood: An initial behavior genetic analysis. Journal of Child Psychology and Psychiatry, 36, 1443–1458. Silberg, J. L., Erickson, M. T., Meyer, J. M., Eaves, L. J., Rutter, M., & Hewitt, J. K. (1994). The application of structural equation modeling to maternal ratings of twins. Journal of Consulting and Clinical Psychology, 62, 510–521. € Sjoberg, R. L., Ducci, F., Barr, C. S., Newman, T. K., Dell’Osso, L., Virkkunen, M., & Goldman, D. (2008). A non-additive interaction of a functional MAOA VNTR and testosterone predicts antisocial behavior. Neuropsychopharmacology, 33, 425–433. Slutske, W. S., Heath, A. C., Dinwiddie, H. H., Madden, P. A. F., Bucholz, K. K., Dunne, M. P., Statham, D., & Martin, N. G. (1997). Modeling genetic and environmental influences in the etiology of conduct disorder: A study of 2,682 adult twin pairs. Journal of Abnormal Psychology, 106, 266–279. Strous, R. D., Bark, N., Parsia, S. S., Volavka, J., & Lachman, H. M. (1997). Analysis of a functional catechol-o-methyltransferase gene polymorphism in schizophrenia: Evidence for association with aggressive and antisocial behavior. Psychiatry Research, 69, 71–77. Taylor, J., McGue, M., & Iacono, W. G. (2000). Sex differences, assortative mating, and cultural transmission effects on adolescent delinquency: A twin family study. Journal of Child Psychology and Psychiatry, 41, 433–440.
Genes, Twin Studies, and Antisocial Behavior
183
True, W. R., Heath, A. C., Scherrer, J. F., Xian, H., Lin, N., Eisen, S. A., Lyons, M. J., Goldberg, J., & Tsuang, M. (1999). Interrelationship of genetic and environmental influences on conduct disorder and alcohol and marijuana dependence symptoms. American Journal of Medical Genetics, 88, 391–397. Vaske, J. (2009). The role of genes and abuse in the etiology of offending. Unpublished dissertation, University of Cincinnati, Cincinnati, OH. Vaske, J., Newsome, J., Makarios, M., Wright, J. P., Boutwell, B. B., & Beaver, K. M. (2009). Interaction of 5HTTLPR and marijuana use on property offending. Biodemography and Social Biology, 55, 93–102. Vaughn, M. G., DeLisi, M., Beaver, K. M., & Wright, J. P. (2009a). DAT1 and 5HTT are associated with pathological criminal behavior in a nationally representative sample of youth. Criminal Justice and Behavior, 36, 1113–1124. Vaughn, M. G., Beaver, K. M., & DeLisi, M. (2009b). A general biosocial paradigm of antisocial behavior: A preliminary test in a sample of adolescents. Youth Violence and Juvenile Justice, 7, 279–298. Widom, C. S., & Brzustowicz, L. M. (2006). MAOA and the ‘‘cycle of violence:’’ childhood abuse and neglect, MAOA genotype, and risk for violent and antisocial behavior. Biological Psychiatry, 60, 684–689. Wright, J. P., & Boisvert, D. (2009). What biosocial criminology offers criminology. Criminal Justice and Behavior, 36, 1228–1240. Young, S. E., Smolen, A., Hewitt, J. K., Haberstick, B. C., Stallings, M. C., Corley, R. P., & Crowley, T. J. (2006). Interaction between MAO-A genotype and maltreatment in the risk for conduct disorder: Failure to confirm in adolescent patients. American Journal of Psychiatry, 163, 1019–1025. Young, S. E., Stallings, M. C., Corley, R. P., Krauter, K. S., & Hewitt, J. K. (2000). Genetic and environmental influences on behavioral disinhibition. American Journal of Medical Genetics, 96, 684–695.
PART III THE BRAIN AND POLITICAL BEHAVIOR
NEUROLOGICAL IMAGING AND THE EVALUATION OF COMPETING THEORIES Dustin Tingley INTRODUCTION In recent years, social scientists have begun exploring the neurological foundations of behavior in an attempt to gain a more complete understanding of decision-making in the realms of both politics and economics (see Cacioppo & Viser, 2003; Fowler & Schreiber, 2008; McDermott, 2009; Caplin & Schotter, 2008). However,1 it is still unclear to many how an approach focused on the brain’s operation can reach beyond a description of the biological process generating some behavior to predicting when and why such behavior occurs. Observing a pattern of brain activity ‘‘x’’ alongside behavior ‘‘z’’ does not in itself give us a better understanding of why ‘‘z’’ happened, or why departures from ‘‘z’’ happen, beyond simply providing a more mechanical description of the process leading to ‘‘z.’’ Even if we identify a sequence of connections from neurological activity and environmental stimulation to cognitive and psychological processes to political behavior, we must consider how much each component contributes to the political behavior under investigation. If the neuroscience is not doing any real work, that is,
Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 187–204 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009010
187
188
DUSTIN TINGLEY
generating specific and testable hypotheses that build on and critique our current understanding of political behavior, then it is hard to see how neuroscience informs political science. If neuroscientific data does provide novel insights, or will develop in a way that will, then political science might have much to learn. There are ample reasons to suspect that neuroscience and related disciplines can help provide a more accurate and expansive understanding of choice behavior of interest to political scientists. In line with an earlier article on the topic, I continue to hold a certain sense of caution in how political science proceeds in the use of neuroscience data for several reasons. While neuroscience and related disciplines continue to make significant progress, these fields continue to have considerable conceptual and methodological debates. Researchers should be aware of how unsettled disputes and problems migrate into their own studies. Assuming (perhaps optimistically) that these debates are irrelevant or of subsidiary importance to political science, there is also a critical question of exactly how to best use neurological imaging data. I advocate a particular approach to using neurological data that emphasizes comparative theoretical testing (Clarke, 2007) via the demonstration of particular neurological mechanisms associated with competing theories of political behavior. If different theories developed by political science suggest different neurological mechanisms, then data on which mechanisms are actually operative, and under what conditions, can be used to help test the competing theories. Such practice would be, I believe, an ideal way for political scientists to directly use neurological data. I structure the remainder of the chapter as follows. First, I review several recent discussions on the use of neuroimaging in economics that helps illustrate the types of debates likely to emerge in political science. The debate will be most helpful if competing theories that make different predictions on what neural mechanisms will be involved are tested. Next, I review a neurological imaging technique, functional magnetic resonance imaging (fMRI),2 and how it is already being used and published by political scientists and economists today. This motivates a discussion on how particular theories within political science might suggest different approaches to using neurological imaging. I argue that understanding the causal role of particular neural mechanisms can help adjudicate between competing theories of behavior. I provide an example of this usage of fMRI from recent work in neuroeconomics as a template for interested political scientists and review an earlier example from the political science literature.
Neurological Imaging and the Evaluation of Competing Theories
189
IMAGING AND NEURAL MECHANISMS IN SOCIAL SCIENCE Why should social scientists be interested in the activity of the brain? Interested political scientists and economists offer similar rationales (e.g., McDermott, 2004, 2009; Camerer, Loewenstein, & Prelec, 2004). First, many political and economic decisions are made automatically without conscious reflection or calculation. This is important for many reasons, but foremost because it calls into question the decision maker postulated in rational choice models.3 Second, emotions influence decisions. A number of prominent studies in political science have implicated the affective system in decisionmaking (e.g., Brader, Valentino, & Suhat, 2008; Marcus, Neuman, & MacKuen, 2000). The common theme here is that many decisions occur at an automatic level until some stimulus is unexpected or novel. The emotional system assists in being able to ‘‘turn on’’ higher-level conscious systems that analyze and deal with the novelty at hand (Marcus et al., 2000; Lieberman, Schreiber, & Ochsner, 2003). Third, the same behavior can be generated by multiple motivations with different neurological mechanisms, which then have different substantive implications.4 In sum, these arguments imply that neurological data can play a role in political science. More recently, the championing of neurological evidence to explain or predict economic decision-making has been challenged by several leading microeconomic theorists. For example, Gul and Pesendorfer (2008) argue that economics is a study of revealed preferences. The process that leads to these preferences simply is not of interest to economics. Insofar as there exist multiple psychological processes that could give rise to the same choice behavior, studying these processes has little bearing on the study of economic choices. This perspective, and various challenges to it, appears in a recently published and fascinating volume that interested political scientists should consult (Caplin & Schotter, 2008). In my experience, there are many political scientists who share Gul and Pesendorfer’s position. Others are less willing to ex ante demarcate what political science is or is not, but hold a ‘‘proof is in the pudding’’ mentality. Below I sketch out how neuroimaging studies can test competing theories that each suggests different neurological mechanisms. If adjudication between competing theories using neuroimaging data is successful, proponents of the Gul and Pesendorfer position will likely have fewer adherents in political science. Given my argument that neuroscience could be useful if it lets us measure the operation of different hypothesized neural mechanisms, I briefly review
190
DUSTIN TINGLEY
how fMRI allows us to observe the operation of different neural mechanisms, by relying on the fact that the brain needs oxygen to function. The oxygen is supplied by blood whose level of oxygenization can be measured. Local changes in brain activity appear to cause local changes in oxygen use. The dominant method of measuring these changes is to monitor the blood-oxygen-level dependent (BOLD) signal through fMRI, which measures the differences in magnetic distortion of excited hydrogen molecules in oxygenated versus deoxygenated blood. This yields relatively precise spatial and temporal resolution of oxygen use in the brain. Critical to understanding the BOLD signal are the physiological mechanisms that govern blood flow and the establishment of a ‘‘baseline’’ rate of oxygen usage from which to compare changes due to stimulus and selected behaviors. The physiological mechanism that produces changes in blood flow predicts that there will be an initial dip in the BOLD signal, followed by a peak above baseline level as the system overcompensates for oxygen usage before returning back to baseline level (Logothetis, 2003; Heeger & Ress, 2002; Beauchamp, 2002). The fMRI technique is neither foolproof nor universally interpreted, and below I discuss some of these limitations. The use of fMRI has been much more common in economics than in politics, perhaps leading to the push back led by Gul and Pesendorfer. Questions about concepts like rewards and bargaining (Montague & Berns, 2002; Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003), advertising (Wells, 2003; Thompson, 2003), and mentalization (the ability to consider the intentions/perceptions of others; Ramnani & Miall, 2004; Blakemore & Decety, 2001; Bhatt & Camerer, 2005) have been studied using fMRI techniques. Social cooperation (Rilling et al., 2002; McCabe, Houser, Ryan, Smith, & Trouard, 2001) and the role of moral judgment on decisionmaking (Greene, Sommerville, Nystrom, Darley, & Cohen, 2001), which are relevant to both economists and political scientists, have also been evaluated using fMRI imaging. Similarly, fMRI imaging in experiments on the influence of race on judgment formation (Phelps & Thomas, 2003; for review, see Ochsner & Lieberman, 2001, pp. 720–721) and political sophistication (Schreiber & Iacoboni, 2004) have yielded interesting results. A recent use of fMRI in economics by Delgado, Schotter, Ozbay, and Phelps (2008) provides a good example of how we might best use fMRI data. A canonical puzzle in economics is the ‘‘Winner’s Curse,’’ where winners of auctions frequently pay more than they should have in equilibrium (Thaler, 1993). One explanation is that individuals are risk averse. As a result, individuals bid more than they should. A second mechanism is that a ‘‘joy in winning’’ motivates individuals. The joy in
Neurological Imaging and the Evaluation of Competing Theories
191
beating other individual bidders drives overbidding as opposed to holding a strong desire to avoid the feeling of losing. A final explanation, and the one ultimately supported by the experimental results, is that contemplation over losses in a social situation drives overbidding. Key here is that individuals especially want to avoid losing against another person. One way to test these different explanations is to deduce different neurological mechanisms that each explanation would suggest as well as experimental designs that put the appropriate environmental conditions into place, such as whether social competition is present. Earlier work (Knutson, Fong, Bennett, Adams, & Hommer, 2003) suggested that one region of the brain associated with losses and gains is the striatum. In the Delgado experiment, both a lottery and auction task were conducted, with only the latter involving any social competition. The authors found a differential response in the right striatum between lottery losses and auction losses, but no such difference between lottery and auction wins. In explaining their results, the authors write: The lack of an enhanced BOLD response in the striatum to wins (in the auction compared with the lottery) suggests that the ‘joy of winning’ may not be mediating overbidding in experimental auctions. In contrast, the stronger BOLD response to losses in the auction game suggests that a fear of losing a social competition may be linked to overbidding. The fear of losing the social competition of an auction may lead to a striatal response similar to that observed in loss aversion. However, because no actual losses occurred in this experiment, it would appear that the ‘fear of losing’ the social competition was a factor independent of pure loss aversion. (p. 1851)
While a follow-up discussion on the merits of Delgado paper suggested some doubts in their findings and the novelty of their preferred explanation of contemplation of losses in a social context (Maskin, 2008), the chapter nicely demonstrates ways to use neurological data to adjudicate between social science theories.
COMPARATIVE THEORY TESTING IN POLITICAL SCIENCE WITH NEUROSCIENTIFIC DATA We have briefly seen how neuroimaging studies can supplement investigations of how people behave politically and economically. Let us now delve into more detail about how different approaches to studying political decision-making suggest different types of neurological evidence. Camerer et al. (2005, p. 16) suggest a heuristic that sees brain operations explained by whether they are controlled or automatic, and whether they are
192
DUSTIN TINGLEY
cognitive or affective (visualized as a 2 2 diagram). Controlled/cognitive behavior is the traditional domain of economic rational choice, where actors consciously attempt to satisfy some utility function that they can explicitly evaluate. For example, one perspective on vote choice assumes that individuals consciously evaluate how a candidate will satisfy their values and material needs. This is a controlled and cognitive process, and, in theory, more likely to activate an executive region of the brain such as the lateral prefrontal cortex (Knight & D’Esposito, 2003). Controlled/affective behavior is what ‘‘methods actors’’ use who consciously elicit emotional reactions. Campaigning politicians who strategically use nonverbal displays (or body language) to elicit emotions provide a concrete example in politics. Automatic/cognitive processes are immediate responses to factual information. Finally, automatic/affective responses include feelings associated with experiences (such as fear or elation) or biological feelings like hunger and sex drive. Quite plausibly, voters might also prefer candidates, in part, because they are physically attractive (Stolberg, 2004; Tingley, 2007; Schubert & Curran, 2001; similarly, see Zebowitz, 1994) or based on other features of a candidate’s appearance (e.g., Ballew & Todorov, 2007; Atkinson, Enos, & Hill, 2009). Because each of these categories may implicate different regions of activation in the brain, or patterns of activation across regions, different types of neurological imaging evidence might be necessarily dependent on what is being studied.
C/X and Affective Intelligence Models of Politics In the Winner’s Curse example discussed above, there were rival theories that sought to explain why auction winners often overbid. To adjudicate between the theories, the authors isolated particular neurological mechanisms in conjunction with different elements of the experimental design. In what follows I construct a contrast between two developing literatures: the ‘‘reflective/reflexive systems’’ model (Lieberman et al., 2003) and the ‘‘Affective Intelligence’’ model (Marcus et al., 2000). While these are related, they suggest slightly different approaches to the role of cognition and affect in human political behavior. In what follows I develop these contrasting view and discuss how neurological imaging information can help adjudicate between the competing perspectives. The comparative testing, I believe, represents a key component of a productive research design linking neuroscience and political science.
Neurological Imaging and the Evaluation of Competing Theories
193
Reflexive and Reflective Processing: The C/X Model Lieberman et al.’s (2003) model posits two different neurocognitive systems: the X-system and the C-system. The X-system (named for the ‘x’ in reflexion), consisting of the lateral temporal cortex, amygdala, and basal ganglia, spontaneously and often nonconsciously integrates current goals, context, perceptions, and activated cognition into a coherent whole that guides the stream of consciousness and current behavior. The C-system (named for the ‘c’ in reflection), consisting of the prefrontal cortex, anterior cingulate cortex, and medial temporal lobes, is recruited when the X-system fails to create coherent outputs from the different sources of input. (p. 689)
According to this model, the X-system guides behavior by incorporating a large range of information (e.g., sensory and motor processing, emotional and social information) until the C-system recognizes a problem in the X-system processes. It then ‘‘interrupts’’ or ‘‘overrides’’ and attempts to resolve the task at hand. In neurophysiological terms that illustrate the complexities of the process, the C-system is monitored by the anterior cingulate, which upon detecting a problem sends a signal to the prefrontal cortex, where a serial processing system resolves the situation. Essentially this is the same distinction between automatic and controlled processes discussed above (Camerer et al., 2005, p. 16), but with a specification of how and why the brain modulates between the two. In their application of this model, Schreiber and Iacoboni (2004) show that, relative to the unsophisticated, politically sophisticated subjects are more likely to have elements of the X-system activated when asked political questions. This is because political ‘‘sophisticates,’’ who have facility with names, dates, or political concepts and have habits of ‘‘associative links formed through extensive learning histories’’ (Lieberman et al., 2003, p. 689), are well accustomed to the processing of political topics. Thus, this model considers the activation/deactivation of reflexive and reflective processes in the brain and how this is mediated by (1) actual familiarity with a set of political topics (which captures the degree of familiarity/novelty of the situation) and (2) emotional feedback that unconsciously provides motivation to ‘‘analyze’’ increasingly unfamiliar situations that generate feelings of unease.
‘‘Affective Judgment’’ Marcus et al. (2000) emphasize the role of affect much more centrally. Their approach builds upon Gray’s model of behavioral approach and behavioral
194
DUSTIN TINGLEY
inhibition.5 ‘‘The Behavioral Approach System gauges the success or failure of recalled actions, contemporary experience, and anticipated activities that fall within the category of previously learned behaviors’’ (Marcus & MacKuen, 2001, pp. 45–46). Emotions implicated by this system range from enthusiasm to depression, and assist in evaluating the success/failure of strategic action. The behavioral inhibition system ‘‘generates moods of calmness and relaxation when the match of incoming sensory signals against anticipated normal execution of plans indicates nothing of concern. It generates moods of increasing nervousness and anxiety when the comparison of environmental information and what would be expected from normal execution of plans indicates a mismatch’’ (p. 47). The combination of the approach and inhibition systems, which are chiefly constituted by the limbic system of the brain, helps individuals navigate the myriad uncertainties of politics.6 While the Affective Intelligence approach to my knowledge has yet to utilize neuroimaging techniques like fMRI, the neuroscience literature it builds from has certainly begun to. Because this approach posits a particular set of neurological processes, fMRI and other imaging techniques will provide a better delineation of what these processes are and how they change under various conditions.
Similarities and Differences The C/X and Affective Intelligence approaches share some significant conceptual similarities. All emphasize the role of habits, automatic responses, and controlled processes. The main difference is that the Affective Intelligence approach more explicitly theorizes the way affective systems modulate and engage controlled decision-making; the approach offered by Lieberman and colleagues is much more focused on information and cognitive processing. Lieberman et al. (2003) do not explicitly theorize how the C system ‘‘detects’’ inconsistencies in the X system beyond specifying that the anterior cingulate is activated. They suggest a combination of (1) logical coherency of the decisions and behaviors being produced by the X-system and (2) dynamic system of feedback with affective processes like anxiety that are generated through interaction with the external environment and mediated through some part of the limbic system. For example, Schreiber and Iacoboni (2004) use this model to explore agree/disagree responses to political and nonpolitical statements. While both models recognize the difference between controlled and automatic processes, cognitive elements modulate the interaction between
Neurological Imaging and the Evaluation of Competing Theories
195
the two more centrally in Lieberman et al.’s account, whereas affective systems do much more work for Marcus and colleagues. Both models leave the interaction between cognitive and affective forces relatively unspecified, perhaps because their interaction is not well understood in the neuroscience literature in the first place (LeDoux, 2000, p. 129). Each also attaches their results to particular conceptions of democratic politics that reflect these differences. Marcus clearly sees the problem of low political participation as a result of too little emotion, the result of a dominant intellectual and cultural perspective that sees ‘‘passion as the enemy of reason.’’ ‘‘(t)he most serious damage is done by continuing to endorse the normative conception of citizenship as a singularly cerebral reflection on justice and the common good’’ (2002, p. 135). Of course this is somewhat qualified by a previous statement: ‘‘to say that negative campaigns and sensationalized presentations by candidates, activists, and the media create the conditions for reason does not mean that we should automatically endorse any use of emotion. The particular circumstances and particular choices must, as with all particulars, be judged on the individual merits’’ (p. 134). The central theme, nevertheless, is that people do not care about politics not because they lack some sort of cognitive ability or exposure, but instead because prevailing notions of democracy discourage the role of affect. This is quite different from the story Schreiber (a co-author of the Lieberman et al.’s paper) tells. Political novices need ‘‘models of how experts connect their values to policy choices’’ (Schreiber & Iacoboni, 2004, p. 8). That is, novices need exposure to the way people familiar with politics, from armchair Sunday morning junkies to D.C. policy makers, connect what they value to how they participate democratically. Thus, political novices ‘‘are not at recess’’ when thinking about politics beyond their immediate social context, and what is needed to help them is some way to familiarize themselves with the cognitive components of the political process.7 A Comparison, A Test? A comparative theory approach to testing these models would examine a particular political choice and record localized neural activity in the brain. The ‘‘Affective Judgment’’ perspective predicts that anxiety is increased when actors are in an unfamiliar context. This anxiety is then postulated to motivate conscious, rational, decision-making, and ‘‘considered judgment.’’ The cognitive nature of the C system in the Lieberman’s model suggests that the role of emotion is less important in the actual ‘‘override’’ process. The brain’s focus, during the override process, is on resolving the cognitive
196
DUSTIN TINGLEY
details of the decision task at hand, and thus difficulty in resolving these details is what drives the ‘‘override’’ process. An example of a study that contrasts the approaches taken in the ‘‘Affective Judgment’’ and ‘‘reflective/ reflexive’’ (C/X) models is Pochon et al. (2002) who investigate reward systems in the brain and the cognitive difficulty of a task. They suggest that with increasing cognitive complexity, the limbic system (which controls affective processes) is ‘‘gated,’’ possibly so as not to interfere (distract) the operation of higher level cognitive areas like the dorsolateral prefrontal cortex. ‘‘We suggest that a dynamic interplay is created between activated cognitive areas necessary to maintain a high level of cognitive performance (the network for WM (working memory) and attention) and affective areas deactivated because they may process counter productive signals interfering with performance’’ (p. 5673). The evidence suggested by Pochon et al. is that a gating mechanism occurs to reduce interference from the paralimbic system on cognitive systems like the dorsolateral prefrontal cortex. This gating process is related to the cognitive complexity of the task and the stakes at hand. The ‘‘override’’ process is initiated by emotional disturbance but then is gated and regulated depending on the cognitive demands of the task. Such a result fits the contours of the C/X model more clearly than it does in the Affective Intelligence perspective, which sees motivation for cognitive reasoning coming from the limbic system. A more defined comparative research program would be needed to more concretely test the theories. The results also suggest need for theorizing about the interrelationship between the stakes and the complexity in decision environments.
LIMITATIONS ON COLLECTING AND INTERPRETING NEUROSCIENTIFIC DATA IN POLITICAL SCIENCE The previous section argued that different ‘‘social’’ science theories could suggest different patterns of neurological activation and highlighted potential shortcomings in current neuroscience models. While the ability to use neurological imaging to discriminate between competing social science theories is exciting, it is important to be cautious about limitations in both the technology and the current state of consensus in the neuroscientific disciplines. In other words, ‘‘conceptual problems (can) migrate’’ between academic disciplines (Johnson, 2002). A more exhaustive discussion is contained elsewhere (Tingley, 2006) and so I provide only a short summary here.
Neurological Imaging and the Evaluation of Competing Theories
197
Modular versus Distributed Information Processing Within neuroscience, there is a continuum of research perspectives on the question of modular (specific functions–specific brain modules) or distributed (specific functions–many brain regions) processing in the brain. While many studies focus on finding the functional properties of distinct neuroanatomical regions, there is conflicting theoretical and empirical evidence of their existence (Cohen & Tong, 2001; for how this surfaces in in vivo and in vitro neurological studies, see Steriade, 2001; Kurzban & Haselton, 2005; Uttal, 2001; for debates regarding the role of the amygdala [of interest to several political scientists], see Adolphs, 2003, p. 169; Baxter & Murray, 2002; Ochsner & Lieberman, 2001, p. 726). A more multifaceted view would study the various functions of specific brain regions, such as the amygdala. Indeed debate about modularity has a very long history (Star, 1989).
(Illusory) Images Obtaining a clean fMRI measurement of neural processes is difficult even with multimillion dollar equipment. Researchers should be cautious in a couple of ways. The measurements of decreases or increases in the BOLD signal are all compared to a ‘‘baseline’’ level that assumes a lack of direct activation from external stimulus. This poses a difficult problem for several reasons. First, it appears that the ‘‘baseline’’ level of neurological activity ‘‘is already tuned to interpreting and categorizing the world as social’’ (Adolphs, 2003, p. 174). Thus, isolating the effect of neural activation must consider how the ‘‘background’’ activities of other regions also contribute to the process at hand. Establishing this ‘‘baseline’’ can be complicated not only by how an experiment is actually set up (in terms of the behavioral category selected as the ‘‘baseline’’) but also by ‘‘task-independent’’ deactivations of particular regions by precise mechanisms that are not fully understood (Gusnard & Raichle, 2001, similarly see Overgaard, 2004). This is exacerbated by the typically small level of fMRI-captured BOLD signals over the baseline (Raichle & Gusnard, 2002; e.g., see Schreiber & Iacoboni, 2004, p. 7). Furthermore, to argue, as some theorists do, that emotions modulate sensory information before the details of political information are processed, Marcus et al. (2000, chap. 4) make an assumption about a baseline level of monitoring that changes depending on feelings of enthusiasm or anxiety. Unfortunately, such theories of politics do not then
198
DUSTIN TINGLEY
specify what normal ‘‘everything makes sense’’ functioning in response to social stimuli looks like versus nonnormal ‘‘this is not what is expected to be happening’’ functioning. Without a better understanding of the contribution of ‘‘baseline’’ activity to normal human functioning in response to diverse situations and types of information, investigations of neural activations will be necessarily vague in terms of what they imply. The BOLD response may vary due to normal aging processes, moreover, complicating the interpretation of results that use subjects that span age groups or extrapolation of studies from one age group to another (D’Esposito, Deouell, & Gazzaley, 2003). Even effects of common substances like caffeine can complicate inferences (Laurienti et al., 2002). Individual variations in brain anatomy pose a challenge for functional location in the brain (Brett , Johnsrude, & Owen, 2002). While significant steps have been made in permitting better localization, a better understanding is required of the functional properties of parts of the brain that are being localized. This becomes increasingly acute as localization moves to higher order functions beyond the primary motor and sensory cortical areas (Way, 2003). Finally, fMRI does have a somewhat limited degree of time resolution, usually between 1 and 4 seconds, due to signal/noise problems that emerge. Additionally, higher temporal resolutions have revealed nonlinearities in the BOLD signal and the normalization of the BOLD signal takes between 16 and 20 seconds, creating similar temporal resolution problems (Pfeuffer, McCullough, Van de Moortele, Ugurbil, & Hu, 2003). While other imaging techniques with better temporal resolution can be used in conjunction with fMRI, such as event-related potential (ERP) measurements, a number of practical issues emerge due to the use of strong magnetic fields. Additional issues include the appropriate statistical procedures to use (Lee, Yoon, Chung, Song, & Park, 2004), analysis design for acceptable statistical significance calculations (Editorial, 2001), and necessity of lesion studies or transcranial magnetic stimulation to rule out spurious relationships in imaging data (Adolphs, 2003, p. 6; Camerer et al., 2005). It warrants restating that these methodological problems can migrate into the evaluation of social scientific theories. Regardless, vast conceptual and technological improvements in the recent past point to similar progress in the future.
CONCLUSION The possibilities for interesting interdisciplinary work between political scientists and neuroscientists are immense. For example, the study of
Neurological Imaging and the Evaluation of Competing Theories
199
coalition formation, with theories ranging from individualistic rational choice to institutional effects, has long been a principal focus for political scientists. Accessing the neurological and psychological dimensions of coalition formation would be a welcome avenue of research. What are the neurological substrates of screening potential partners? As coalitions become more salient, are there changes in how the brain monitors possible disruptions/deviations? Is the decision by someone with political experience to join a coalition identical in process to what a political novice would use? More generally, the prospect of social cooperation (such as in the prisoner’s dilemma) is a central theme in political science. The political decisionmaking underlying coalition formation and cooperation involves (to name just a few) trust, monitoring, reputation, reciprocation, evaluation of the intentions of others, and estimation of payoffs/rewards. Neuroscientific work that probes these traditionally central concepts is illustrated by a recent study of altruistic norm enforcement (de Quervain et al., 2004). Both political scientists and neuroscientists stand to gain from working together to better explicate many crucial features of modern political science, just as increasing numbers of economists have begun to explore neuroscientific methods. But such collaborations will be most fruitful if they are done in an environment that tests theories comparatively. If competing explanations of political behavior suggest different neurological mechanisms, then fMRI will be particularly useful. Our understanding of the human brain, and its relation to higher levels of explanation in the study of humans, will come from both the development of theory and the collection of evidence. fMRI and related technologies give political scientists and economists an expansive new source of raw data. However, just as a failure to focus on conceptual, as opposed to empirical, progress undermines political science ‘‘pre-neuropolitics’’ (Johnson, 2002, 2003), similar problems apply to the development of ‘‘neuropolitics.’’ To this end, political scientists, economists, and neuroscientists need to work with each other in ways that explicitly acknowledge limitations in their conceptual framework and measurement techniques and hence preventing alarms about the role of neuroscience in the study of politics (e.g., Wahlberg, 2004; Editorial, 2003, 2004). Given that political science is currently discussing the merits of different ‘‘scientific’’ approaches to political science, such as rational choice theory, neuroscience may be able to provide fresh perspectives on seemingly incommensurable research traditions. Lupia’s (2002) efforts to bring formal modeling of psychological processes into political explanation, Bueno de Mesquita and McDermott’s (2004) recent statement on prospect theory, and
200
DUSTIN TINGLEY
Blank’s (2003a) discussions on public policy all suggest that the ground is fertile for novel approaches to inquiry in political science. Politics offers an incredibly rich source of information about human emotion, cognition, and behavior: alliance formation, provision of public goods, use of force, and partisanship, to name just a few. Further scientific exploration of these and other political behaviors should tread slowly yet surely.
NOTES 1. A substantially earlier version of this chapter appears in Tingley (2006). 2. I do not focus on other imaging technologies, such as positron emission topography (PET), because of the growing frequency that fMRI is being used (PET requires the administration of a radioactive substance into the bloodstream) and because many of my general arguments will cross apply. 3. Of course, rational choice theory does not specify that people must be ‘‘conscious’’ or in ‘‘control’’ of their decision-making process in order to be rational, or even modeled as being rational (Ordeshook, 1995, p. 178). However, rational choice theory often used in explanations where the posited actors are conscious and acting intentionally. This stems from the intentionalist form of explanation being used (Tingley, 2007; Elster, 1983). Of course, the neuroscience program explored here is trying to build those mechanisms instead of relegating them to the black box that rational choice theorists purportedly try to avoid (Boudon, 1998). 4. Camerer (2003a, 2003b), citing Romer (2000), give the example of two people who do not eat peanuts. One person likes the taste of peanuts but consciously does not eat them because of an allergy. The other person had an adverse experience with peanuts but can still eat them; not eating peanuts is a habit formed from a harsh memory. Each method of explaining not eating peanuts suggests different economic consequences like differences in price elasticity. Darren Schrieber and colleagues suggest a similar example from political science: responses to political questions can be habit based due to a high degree of familiarity with the issue (sophistication) or the result of a highly conscious effort to construct a response, given a very low level of familiarity with the subject (Lieberman et al., 2003; Schreiber & Iacoboni, 2004). 5. Interestingly, Marcus and MacKuen (2001) find that the third piece of Gray’s model, relating to ‘‘fight–flight’’-type interactions, is ‘‘of limited application’’ to the study of politics, presumably because of the primitiveness that such a situation involves (p. 44). Morikawa, Hanley, and Orbell’s (2002) analysis of cognitive requirements in ‘‘hawk-dove’’ games paints a more central picture for ‘‘fight-flight’’ situations in political interaction. 6. Another well-developed theory, which I do not cover here only for simplicity’s sake, is known as the ‘‘hot-cognition’’ model. It argues, among other things, that political concepts are stored in memory with effective associations attached to them. Thus, exposure to these concepts generates automatic affective impressions upon recall (Morris, Squires, Taber, & Lodge, 2003; Lodge & Taber, 2000). For a general review of the role of emotions in politics, see Marcus (2000, 2002, 2003).
Neurological Imaging and the Evaluation of Competing Theories
201
7. Schreiber acknowledges that such a reading of his work is plausible, though he maintains more of an agnostic stance here: both effective and cognitive forces can plausibly modulate responses to political ‘‘facts’’ and arguments. Further, how both of these systems do this is not well understood, largely because much of the study of effect in relation to politics is too oriented toward the study of disaffiliative effect like fear and anxiety (Schreiber, personal conversation).
REFERENCES Adolphs, R. (2003). Cognitive neuroscience of human social behavior. Nature Reviews: Neuroscience, 4, 165–178. Atkinson, M. D., Enos, D., & Hill, S. (2009). Candidate faces and election outcomes: Is the face-vote correlation caused by candidate selection? Quarterly Journal of Political Science, 4(3), 229–249. Baxter, M., & Murray, E. (2002). The amygdala and reward. Nature Reviews Neurosciences, 3(7), 563–573. Beauchamp, M. (2002). Functional MRI for beginners. Nature Neuroscience, 5(5), 397–398. Ballew, C., & Todorov, A. (2007). Predicting political elections from rapid and unreflective face judgments. Proceedings of the National Academy of Sciences, USA, 104(46), 17948–17953. Bhatt, M., & Camerer, C. (2005). Self-referential thinking and equilibrium as states of mind in games: fMRI evidence. Games and Economic Behavior, 52(2), 424–459. Blakemore, S., & Decety, J. (2001). From the perception of action to the understanding of intention. Nature Reviews Neurosciences, 2, 561–568. Blank, R. (2003a). Policy implications of advances in cognitive neuroscience. American Association for the Advancement of Science: Science and Technology Policy Yearbook. Available at www.aaas.org/spp/yearbook/2003/ch6.pdf. Boudon, R. (1998). Social mechanisms without black boxes. In: P. Hedstrom & R. Swedberg (Eds), Social mechanisms: An analytical approach to social theory (pp. 172–203). Cambridge, MA: Cambridge University Press. Brader, T., Valentino, N., & Suhat, E. (2008). What triggers public opposition to immigration? Anxiety, group cues, and immigration. American Journal of Political Science, 52(4), 959–978. Bueno de Mesquita, B., & McDermott, R. (2004). Crossing no man’s land: Evidence from the trenches. Political Psychology, 25(2), 271–287. Cacioppo, J., & Viser, H. (2003). Political psychology and social neuroscience: Strange bedfellows or comrades in arms? Political Psychology, 24(4), 647–656. Camerer, C., Loewenstein, G., & Prelec, D. (2004). Neuroeconomics: Why economics needs brains. Scandinavian Journal of Economics, 106(3), 555–579. Camerer, C., Loewenstein, G., & Prelec, D. (2005). Neuroeconomics: How neuroscience can inform economics. Journal of Economic Literature, 43(1), 9–64. Camerer, C. F. (2003a). Strategizing in the brain. Science, 300, 1673–1675. Camerer, C. F. (2003b). Behavioral studies of strategic thinking in games. Trends in Cognitive Sciences, 7(5), 225–231. Caplin, A., & Schotter, A. (2008). The foundations of positive and normative economics. New York: Oxford University Press.
202
DUSTIN TINGLEY
Clarke, K. (2007). The necessity of being comparative: Theory confirmation in quantitative political science. Comparative Political Studies, 40(7), 886–908. Cohen, J., & Blum, K. (2002). Reward and decision. Neuron, 36, 193–198. Cohen, J. D., & Tong, F. (2001). The face of controversy. Science, 293, 2405–2407. Delgado, M., Schotter, A., Ozbay, E., & Phelps, E. A. (2008). Understanding overbidding: Using the neural circuitry of reward to design economic auctions. Science, 321(5897), 1849–1852. D’Esposito, M., Deouell, L., & Gazzaley, A. (2003). Alternations in the BOLD fMRI signal in ageing and disease: A challenge for neuroimaging. Nature Reviews Neuroscience, 4(November), 863–872. De Quervain, D., Fischbacher, U., Treyer, V., Schellhammer, M., Schnyder, U., Buck, A., & Fehr, E. (2004). The neural basis of altruistic punishment. Science, 305, 1254–1258. Editorial. (2001). Analyzing functional imaging studies. Nature Neuroscience, 3(4), 333. Editorial. (2003). Scanning the social brain. Nature Neuroscience, 6(12), 1239. Editorial. (2004). Brain scam? Nature Neuroscience, 7(7), 683. Elster, J. (1983). Explaining technical change. New York: Cambridge University Press. Fowler, J., & Schreiber, D. (2008). Biology, politics, and the emerging science of human nature. Science, 322(5903), 912–914. Greene, J. D., Sommerville, R. B., Nystrom, L. E., Darley, J. M., & Cohen, J. D. (2001). An fMRI investigation of emotional engagement in moral judgement. Science, 293, 2105–2108. Gul, F., & Pesendorfer, W. (2008). The case for mindless economics. In: A. Caplin & A. Shotter (Eds), The foundations of positive and normative economics. New York: Oxford University Press. Gusnard, D., & Raichle, M. (2001). Searching for a baseline: Functional imaging and the resting human brain. Nature Reviews Neuroscience, 2(October), 685–694. Heeger, D., & Ress, D. (2002). What does fMRI tell us about neuronal activity? Nature Reviews Neuroscience, 3(February), 142–152. Johnson, J. (2002). How conceptual problems migrate: Rational choice, interpretation, and the hazards of pluralism. Annual Review of Political Science, 5, 233–248. Johnson, J. (2003). Conceptual problems as obstacles to progress in political science: Four decades of political culture research. Journal of Theoretical Politics, 15(1), 87–115. Knight, R., & D’Esposito, M. (2003). Lateral prefrontal syndrome: A disorder of executive control. Cambridge, MA: Bradford Books. Knutson, B., Fong, G. W., Bennett, S., Adams, C., & Hommer, D. (2003). A region of mesial prefrontal cortex tracks monetarily rewarding outcomes: Characterization with rapid event-related fMRI. NeuroImage, 18, 263–272. Kurzban, R., & Haselton, R. (2005). Making hey out of straw: Real and imagined controversies in evolutionary psychology. In: J. Barlow (Ed.), Missing the revolution: Darwinism for social scientists. New York: Oxford University Press. Laurienti, P.Laurienti, P. J., Field, A.S., Burdette, J. H., Maldjian, J.A., Yen, Y.F., & Moody, D. M. (2002). Dietary caffeine consumption modulates fMRI measures. NeuroImage, 17, 751–757. LeDoux, J. (2000). Cognitive-emotional interactions: Listen to the brain. In: R. D. Lane & L. Nadel (Eds), Cognitive neuroscience of emotion. New York: Oxford University Press. Lee, S., Yoon, H. W., Chung, J., Song, M., & Park, H. W. (2004). Analysis of functional MRI data based on the hemodynamic response in the human brain. Journal of Neuroscience Methods, 139(1), 91–98.
Neurological Imaging and the Evaluation of Competing Theories
203
Lieberman, M., Schreiber, D., & Ochsner, K. (2003). Is political cognition like riding a bicycle? How cognitive neuroscience can inform research on political thinking. Political Psychology, 24(4), 681–704. Lodge, M., & Taber, C. (2000). Three steps toward a theory of motivated political reasoning. In: A. Lupia, M. McCubbins & S. Popkin (Eds), Elements of reason: Cognition, choice, and the bounds of rationality (pp. 183–213). Cambridge, MA: Cambridge University Press. Logothetis, N. (2003). The underpinnings of the BOLD functional magnetic resonance imaging signal. The Journal of Neuroscience, 23(10), 3963–3971. Lupia, A. (2002). Who can persuade whom? Implications from the nexus of psychology and rational choice theory. In: J. Kuklinski (Ed.), Thinking about political psychology (pp. 51–88). New York: Cambridge University Press. Marcus, G. (2000). Emotions in politics. Annual Review of Political Science, 3, 221–250. Marcus, G. (2002). The sentimental citizen: Emotion in democratic politics. University Park, PA: Penn State University Press. Marcus, G. (2003). The psychology of emotion and politics. In: D. Sears, L. Huddy & R. Jervis (Eds), Oxford handbook of political psychology. New York: Oxford University Press. Marcus, G. E., & MacKuen, M. (2001). Emotions and politics: The dynamic functions of emotionality. In: J. H. Kuklinski (Ed.), Citizens and politics (pp. 41–67). New York: Cambridge University Press. Marcus, G., Neuman, W., & MacKuen, M. (2000). Affective intelligence and political judgment. Chicago: University of Chicago Press. Maskin, E. (2008). Can neural data improve economics?. Science, 321, 1788–1789. McCabe, K., Houser, D., Ryan, L., Smith, V., & Trouard, T. (2001). A functional imaging study of cooperation in two-person reciprocal exchange. Proceedings of the National Academy of Sciences, USA, 98(20), 11832–11835. McDermott, R. (2004). The feeling of rationality: The meaning of neuroscientific advances for political science. Perspectives on Politics, 2(4), 691–706. McDermott, R. (2009). Mutual Interests: The case for increasing dialogue between political science and neuroscience. Political Research Quarterly, 62(3), 571–583. Montague, P., & Berns, G. (2002). Neural economics and the biological substrates of valuation. Neuron, 36, 265–284. Morikawa, T., Hanley, J., & Orbell, J. (2002). Cognitive requirements for Hawk-Dove games: A functional analysis for evolutionary design. Politics and the Life Sciences, 21(1), 2–10. Morris, J., Squires, N., Taber, C., & Lodge, M. (2003). Activation of political attitudes: A psychophysiological examination of the hot cognition hypothesis. Political Psychology, 24(4), 727–745. Ochsner, K., & Lieberman, M. (2001). The emergence of social cognitive neuroscience. American Psychologist, 56(9), 717–734. Ordeshook, P. (1995). Engineering or science: What is the study of politics?. In: J. Friedman (Ed.), The rational choice controversy. New Haven, CT: Yale University Press. Overgaard, M. (2004). Confounding factors in contrastive analysis. Synthese, 141, 217–231. Pfeuffer, J., McCullough, J. C., Van de Moortele, P. F., Ugurbil, K., & Hu, X. (2003). Spatial dependence of the nonlinear BOLD response at short stimulus duration. NeuroImage, 18, 990–1000. Phelps, E., & Thomas, L. (2003). Race, behavior, and the brain: The role of neuroimaging in understanding complex social behaviors. Political Psychology, 24(4), 747–758.
204
DUSTIN TINGLEY
Pochon, J., Levy, R., Fossati, P., Lehericy, S., Poline, J., Pillon, B., Le Bihan, D., & Dubois, B. (2002). The neural system that bridges reward and cognition in humans: An fMRI study. Proceedings of the National Academy of Sciences, USA, 99(8), 5669–5674. Raichle, M. E., & Gusnard, D. A. (2002). Appraising the brain’s energy budget. Proceedings of the National Academy of Sciences of the United States of America, 99(16), 10237–10239. Ramnani, N., & Miall, R. (2004). A system in the human brain for predicting the actions of others. Nature Neuroscience, 7(1), 85–90. Rilling, J. k., Gutman, D. A., Zeh, T. R., Pagnoni, G., Berns, G. S., & Kilts, C. D. (2002). A neural basis for social cooperation. Neuron, 35, 395–405. Romer, J. (2000). Thinking and feeling. American Economic Review, 90(2), 439–443. Sanfey, A., Rilling, J., Aronson, J., Nystrom, L., & Cohen, J. (2003). The neural basis of economic decision making in the ultimatum game. Science, 300, 1755–1758. Schreiber, D., & Iacoboni, M. (2004). Evaluating political questions: Neural systems and functional mechanisms. Paper presented at the 2004 Political Methodology meeting Stanford University. Schubert, J., & Curran, M. (2001). Appearance effects in political careers: Do politicians with good genes get more votes. Paper presented at the Human Behavior and Evolution Society Meetings, London, England. Star, S. (1989). Regions of the mind: Brain research and the quest for scientific certainty. Stanford, CA: Stanford University Press. Steriade, M. (2001). The intact and sliced brain. Cambridge, MA: MIT Press. Stolberg, S. (2004). Cute, sure, but is he electable? The New York Times, July 11. Thaler, R. (1993). The winner’s curse: Paradoxes and anomalies of economic life. Princeton, NJ: Princeton University Press. Thompson, C. (2003). There’s a sucker born in every medial prefrontal cortex. The New York Times, October 26. Tingley, D. (2006). Neurological imaging as evidence in political science: A review, critique, and guiding assessment. Social Science Information, 45(1), 5–33. Tingley, D. (2007). Evolving political science: Biological adaptation, rational action, and symbolism in political science. Politics and Life Sciences, 25(1), 23–41. Uttal, W. (2001). The new phrenology: The limits of localizing cognitive processes in the brain. Cambridge, MA: MIT Press. Wahlberg, D. (2004). Advertisers probe brains, raise fears. The Atlanta Journal-Constitution, February. Way, B. (2003). Topography of the serotonin transporter throughout the prefrontal cortex of the vervet monkey. Ph.D. dissertation in Neuroscience, UCLA. Wells, M. (2003). In search of the buy button. Forbes, September. Available at http:// www.forbes.com/forbes/2003/0901/062.html Zebowitz, D. (1994). Facial maturity and political prospects: Persuasive, culpable, and powerful faces. In: R. Schank & E. Langer (Eds), Beliefs, reasoning, and decision making: Psychologic in honor of Bob Abelson. Hillsdale, NJ: Lawrence Erlbaum Associates.
BRAIN SCIENCES AND POLITICS: SOME LINKAGES Robert H. Blank Like genetics and stem cell research, neuroscience1 promises to be a highly controversial political issue. However, while the political ramifications of genetic research have been well documented and widely analyzed, there has been little systematic scrutiny given to neuroscience. The array of techniques and strategies for intervention in and imaging of the brain are expanding rapidly and are certain to be joined in the future by even more extraordinary capabilities. In addition to treating neural diseases and disorders, these innovations promise increasingly precise and effective means of predicting, modifying, and controlling behavior. Moreover, while advances in neuroscience and technology are impressive, much of the popular literature trends to oversimplify and exaggerate the claims of presumed efficacy, thus heightening the fears of those persons who see it as a threat and creating unsubstantiated hope in those persons who might benefit from it. While this helps to bring attention to neuroscience, it also intensifies the debate and hardens divergent predispositions on what should be done. The 1990s ‘‘Decade of the Brain’’ stimulated neuroscience research on many fronts and resulted in considerable advancement in the science. Unfortunately, we have been slow to develop a political dialogue to anticipate and deal with the enormous implications of the brain sciences. Simply put, our political and social institutions have not kept pace with these advances. At the base, political issues center on how we interpret the implications of these developments, particularly given the complexity of the Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 205–229 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009011
205
206
ROBERT H. BLANK
subject and the speculative nature of much of the evidence to date. The questions discussed here go to the heart of the major problems facing the world today and challenge basic postulates across the political/ideological spectrum as well as the standard social science model of political behavior. As a result, the findings of neuroscience are open to ideological maneuvering designed to put the most expedient spin on their meaning. Although we know considerably more about the brain than we did several years ago, in many ways it is still rudimentary. There are many theories and suppositions, as well as a proliferation of information and findings, as research to identify the anatomical connections and to understand the biochemical, molecular, and genetic mechanisms of the brain accelerates. The heightened potential for ever more precise and effective means to predict, modify, and potentially control behavior raises many political issues. It is now time to broaden the dialogue over how to use these new tools for understanding human behavior. After briefly describing the range of innovations in neuroscience and discussing generic political issues, this chapter summarizes the policy implications of neuroscience research and technologies for our understanding of the human condition. It then speculates on what this all means in terms of politics.
TECHNOLOGICAL CONTEXT As one of the most dynamic and consequential areas of biomedical research, neuroscience must be analyzed in a broader political context. Research initiatives, individual use, and aggregate social consequences of unfolding knowledge about the brain and the accompanying applications require particularly close scrutiny because of the centrality of the brain itself to human behavior and thoughts. As one of the last frontiers of medicine, neuroscience has strong support because it promises to benefit many patients suffering from an array of behavioral, neurological, and mental disorders and injuries. Given the inevitability of expanded strategies for exploration and therapy of the brain, it is important that the political issues surrounding their application be clarified and debated before such techniques fall into routine use. Historically, experimental and clinical interventions in the brain have elicited controversy from many directions. Illuminating are the issues surrounding early innovations of frontal lobotomies, electroshock therapy, and abuses of psychotropic drugs. Although new advances promise considerable benefits, like genetic technologies, the transformation in
Brain Sciences and Politics: Some Linkages
207
neuroscience threatens social values concerning personal autonomy and rights and, for some critics, raises the specter of mind control and a Clockwork Orange-type society. The brain sciences, then, represent an emerging policy arena that must be studied much like genetics, reproductive techniques, and organ transplantation. In fact, given its centrality to human existence, neuroscience policy might well be considered a prototype for analysis of the social impacts of future biomedical interventions. As such, neuroscience policy and politics are a critical area of study for neuroscientists, social scientists, and ethicists. Anxieties over brain intervention reflect the complexity of the object examined – the brain – and the ramifications of any evidentiary path proceeding from its study. Findings in this field do not simply accumulate; they interact, synergistically and convergently, bringing us ever nearer to the very seat of instinct, intellect, emotion, behavior, responsibility, and culpability. Illustrative of neuroscience’s progress has been the visualization and the quantification of mental activity itself and the real-time correlation of that activity with prompted observable behaviors. The potential for progressively more reliable means to predict, to modify, and even to control behavior raises intriguingly vexing questions along the political–philosophical line, but pressingly practical questions along bioethical, jurisprudential, and science-policy lines (Blank, 1999, 2006). To date, interest within the bioethics community, where an active neuroethics subfield has emerged, thus far has exceeded interest within the policy-making and policy-sciences communities (see Giordano & Gordijn, 2010; Racine, 2010; Farah, 2010). As illustrated in Table 1, brain intervention techniques run the gamut from techniques that act directly on the brain including neural grafting, to psychotropic drugs and biologics, to nonevasive virtual reality (VR) and neural imaging, and neurogenetics research. This chapter will focus on direct techniques, pharmaceuticals, and VR. It will also discuss neural imaging where appropriate, but limit its coverage since it is covered in detail in other chapters. Especially provocative has been direct brain intervention, ranging from electroconvulsive therapy (ECT), administered through the scalp, to electronic stimulation of the brain (ESB), wherein impulses are sent and received through insulated wires implanted in targeted regions of the brain. More recent developments include transcranial magnetic stimulation (TMS), which utilizes powerful electromagnetic discharges to alter brain activity, and deep brain stimulation (DBS), where electrodes are implanted in the brain to deliver high-frequency electrical stimulation to treat Parkinson’s disease, epilepsy, and obsessive disorders (Bell, Mathieu, & Racine, 2009;
208
ROBERT H. BLANK
Table 1.
Techniques of Brain Intervention.
1. Direct brain techniques Electroconvulsive therapy (ECT) Transcranial magnetic stimulation (TMS) Electronic brain stimulation (ESB) Brain implants Deep brain stimulation (DBS) Vagus nerve stimulation (VNS) Transcutaneous electrical nerve stimulation (TENS) Transcranial direct current stimulation (tDCS) Magnetic seizure therapy (MST) Psychosurgery 2. Pharmaceuticals and biologics Antipsychotic Antidepressant Antianxiety Hormonal treatment Nootropics (performance enhancers) 3. Virtual reality 4. Brain imaging 5. Neural grafting 6. Neurogenetics (imaging genetics)
Schmetz & Heinemann, 2010). Other applied techniques include vagus nerve stimulation (VNS), transcutaneous electrical nerve stimulation (TENS), transcranial direct current stimulation (tDCS), and magnetic seizure therapy (MST). The more primitive forms of psychosurgery, such as frontal lobotomies, have been replaced by stereotaxic surgery, which destroys tissue in precise spots of the brain using lasers, radiation, cryoprobes, or focused ultrasound beams. A rapidly expanding and provocative area of research involves brain, or neural, implants that connect directly to a person on the surface of the brain or in the neocortex. The major impetus for brain implants comes from research to circumvent those areas of the brain that became dysfunctional after a stroke or other head injuries or implanted pacemaker-like devices to treat mood disorders (Reuters, 2005). Brain implants electrically stimulate, block, or record impulses from single neurons or groups of neurons where the functional associations of these neurons are approximately known. Advanced research in brain implants involves the creation of interfaces between neural systems and computer chips. Implants involving DBS and VNS have become more routine for patients with Parkinson’s disease and
Brain Sciences and Politics: Some Linkages
209
clinical depression, respectively, but more exotic optogenetic implants composed of electrodes or fiber optic wires on the brain’s surface will beam light pulses to either control brain cells or reroute brain activity. Moreover, by introducing a few genes to specific neuron clusters, the cells can be made sensitive to certain wavelengths that can make them fire or shut them off (Dillow, 2010). Drugs to modulate or enhance specific behavioral traits have long since appeared and are proliferating. New generations of antipsychotic, antianxiety, and especially antidepressant drugs that inhibit serotonin reuptake – prototypically fluoxetine (Prozac) – are now among the most widely used pharmaceuticals. Moreover, nootropics, ‘‘brain-performance enhancing drugs’’ or ‘‘smart drugs,’’ are becoming more widely used and are expected to create a huge market in the near future (Hall, 2003; Rose, 2002). For instance, beta-adrenergic blocking agents have long since been used by musicians, many of them players of wind instruments, to blunt stage fright and boost performance. Other enhancers, some hormonally active, have also begun to be used by many professionals. At once more remote and more compelling in its implications is neurogenetics. Promising research areas involve supplementing or blocking neurotransmitters, identifying genes that prevent normal brain development or produce progressive brain degeneration, replacing deleted or defective genes, and removing inoperable brain tumors using viral vectors. Distinct in method is a final class of interventions: operant conditioning by VR, usually with pharmacological augmentation, to treat phobias and posttraumatic stress syndrome (PTSS). VR is a computer-mediated, multisensory technology designed to bring the mind into an alternate reality. VR produces a deliberate effect that is quasi-hallucinatory and has been compared to schizophrenia. As it advances to encompass more senses and to generate more nearly complete perceptual fields less intruded upon by delivery artifact, VR might for some subjects make everyday distinctions between reality and fantasy genuinely and disturbingly difficult. Putative risks of placement in an alternate reality, plausibly including interference with normal psychological processes, disinclination to deal with challenging life circumstances, and even addiction accompanied by secondary social withdrawal, have not attracted much attention, possibly because ‘‘Internet addiction’’ and compulsive electronic gaming are worrisome enough to preempt discussion of more exotic analogues. Antecedent to and concurrent with advances in brain intervention have been advances in our understanding of structural and functional links to human behavior. Noninvasive functional imaging is making analysis of the
210
ROBERT H. BLANK
brain an obligatory intermediary between genotype and behavior (Hurley & Taber, 2008). Until recently, our understanding of brain structure and function was restricted, based on inference from trauma, disease, and autopsy, on extrapolation from animal studies, or on the outcomes of very risky ‘‘therapeutic’’ physical intrusions in patients. Although safe, practical ways to study the living brain functionally began to evolve long ago with electroencephalography, in the mid-twentieth century, radioisotope scanning began to be used. In the 1990s, axial tomography became computerized and then adapted from X-radiation to magnetic resonance imaging (MRI) and positron emission tomography (PET) scanning. Today functional MRI and PET are becoming widely used in behavioral research (see other chapters in this book). Scientists also have developed computer programs that can alter or rearrange anatomical brain images from MRI and PET to match a standardized brain map, thus making it possible to compare the anatomy and function of different brains. In 2003, for example, the Brain Atlas, which is comprised of digitalized high-definition structural maps collected from MRI scans of over 7,000 subjects was published on the Internet. Layered over the anatomical maps are brain functions such as memory, emotion, and language. Users can look at individual pictures, composite pictures of subgroups by age, race or gender, or the composite of all the subjects (Beasley, 2003). In addition to these imaging technologies, other methods have been developed to study brain activity. Chance et al. (1998), for instance, use functional near-infrared spectroscopy (fNIR) that are portable and less expensive than imaging techniques and do not require immobilization of the subject. A related technology, touted by its inventor Lawrence Farwell as forensic brain fingerprinting, is the multifaceted electroencephalographic response analysis (MERA) that measures a specific electrical brain wave known as a P300 (Brain Fingerprinting Laboratories, 2003). A P300 is emitted by the brain within a fraction of a second when an individual recognizes an incoming stimulus as significant. In contrast, irrelevant stimuli do not trigger a P300; thus, by carefully presenting a mixture of significant and insignificant stimuli to a person, memory can be ‘‘detected.’’ Advances in molecular biology, brain imaging, genetic epidemiology, and developmental psychopathology provide a unique opportunity to explore the interplay of genes, brain, and behavior within a translational research framework (Viding, Williamson, & Hariri, 2006). The innovative field of imaging genetics uses brain imaging methods to assess the impact of genetic variation on the human brain (Kempf & Meyer-Lindenberg, 2006). Generally, multiple imaging methods are used in combination to achieve
Brain Sciences and Politics: Some Linkages
211
an optimal characterization of structural–functional parameters in large groups of individuals whose genotypes are then statistically related to these data across the subjects, a form of genetic association study. Although this technique is in its infancy, the emerging literature shows that it can be useful in identifying neural processes involved in mediating the effect of genetic polymorphisms on mental disease risk, thus moving us closer to understanding the complexities of the specific mechanisms involved in the etiology of psychiatric disease. To date studies have dealt with genes involved in risk for schizophrenia, Alzheimer’s disease, and depression, anxiety, and violence, but extensions to genes linked to genetically based behavioral tendencies will likely follow in the near future as our understanding of genetics and the brain improve. Tairyan and Illes (2009) detail the ethical concerns accentuated by imaging genetics.
ETHICS, POLITICS, AND BRAIN INTERVENTION With our every intervention and every observation in every human brain, we face a multitude of ethical and political issues. Even ECT, used more or less successfully since the 1930s, is still ‘‘one of the most controversial treatments in medicine’’ (Carney & Geddes, 2003). The reasons are familiar. Less widely appreciated, though, is how easily traditional topics in clinical and research ethics merge into new topics in political ethics and, therefore, policy. The following are but a sample of the many issues that broach both ethics and politics. First, however, conscientiously elicited, consent is unlikely to be informed fully for any first intervention and may be less competently offered, albeit better ‘‘informed,’’ for subsequent interventions. Consent is of concern prior to any medical procedure, but consent in this case has to come from the very organ that needs evaluation, assistance, or repair or that presents itself, once € or repeatedly, for experimentation (Muller & Walter, 2010). On the one hand, the question arises as to whether fully informed consent is possible; on the other, paternalistic imputation of consent may seem sufficiently sensible to weaken safeguards and trigger fears of potential abuse. Second, while often a fine one in medicine generally, the line dividing experimentation and therapy disappears here in the sense that each person’s brain is unique in ways affecting identity, memories, judgments, and preferences. As the brain defines the person, even minor alterations might shade a definition of self or, more observably, an assessment of character. Moreover, the complexity of the human brain assures side effects that can
212
ROBERT H. BLANK
never be fully predictable. As in other areas of medicine there has been a tendency to move quickly from experimental to therapeutic status in developing these innovations. Although this is understandable given the vulnerability of patients who normally undergo these techniques, their very vulnerability warrants a cautious approach. This concern over risks and uncertainties is especially critical when the techniques are used in nonmedical, especially commercial settings. Caution is especially imperative when an intervention is irreversible, as is the case of psychosurgery, but caution is also important when short- or long-term consequences may be too subtle to notice, too idiosyncratic to define, or too safe within the range of normal to be called adverse. For instance, there is increased concern over the long-term effects of DBS even as it has proliferated as an accepted therapy (Schlaepfer, Lisanby, & Pallanti, 2010). Concern is deepened when treating behavioral disorders without a proven organic origin or when modifying questionably disordered behaviors that are troublesome principally to families, societies, governments, or insurers. Third, any form of intrusion in the brain conjures the image of mind control, thus unsettling assumptions of individual responsibility and widening the scope of culpability. As the seat of personal autonomy and identity, the brain enjoys special status and altering it even slightly raises concerns of manipulation. In parallel fashion, the incremental elucidation of neurological factors in human behaviors, especially aberrant behaviors, may greatly complicate moral, civil, and jurisprudential argument (Gazzaniga, 2005). Fourth, to whatever extent judgments of character are supplanted by judgments of genetics or neurons, stigmatization will adapt, moving from an evaluative realm in which intuition serves imperfectly to one in which it serves only to harm. People with mental disorders are treated differently than those with somatic diseases, and discrimination against the mentally ill, including discrimination in budgeting for their care, is widespread even among the well meaning. In a coming era of neurosciences insight, such discrimination might lessen in old ways while worsening in new ones, the balance of the two trends hard to estimate prospectively. Complicating matters, Meyer-Lindenberg et al. (2008) suggest that frequent false positives are likely in imaging genetics. The potential to pigeonhole and discriminate on the basis of test results could lead to negative consequences including the development of a ‘‘neuroscientific underclass’’ denied access to societal benefits on the basis of their neuroscience test results (Garland, 2004). Fifth, intervention in the brain raises questions of distributive justice, mainly involving the access to services, many of which are costly and are
Brain Sciences and Politics: Some Linkages
213
likely to become even more so. Although resource allocation questions, to date, have been rare, they are critical because while resources are finite, demands and expectations fueled by new technologies have few bounds. While it might be premature to speculate about the relative costs and benefits of yet undeveloped procedures, it is logical to assume that cumulatively their cost will be high. As such, these interventions also raise questions as to who should get priority for use of these procedures, whether insurance companies must pay for what are at best experimental treatments, and to what extent a government can or should regulate their use. Will access be equitable and coverage universal, and, if so, how will it be funded? Or, will it be available to the affluent, but largely denied to persons who lack sufficient resources? This question of access becomes of critical importance as we move into an era in which endowment of social advantage becomes possible through enhancement of brain function. Sixth, and more broadly, what priority should the search for knowledge about the brain and ever expanding uses of this knowledge have vis-a-vis other strategies and health care areas? What benefits will it hold for the population as a whole, compared to other spending options? On the one hand, over the last several decades there has been a proclivity to develop and widely diffuse expensive curative techniques without first critically assessing their overall contribution to health outcomes. On the other hand, the availability of effective and inexpensive imaging tests might provide valuable information for disease prevention and health promotion by targeting individuals who are at heightened risk for diseases that could be reduced by early intervention. What priority should these interventions have then, particularly when used to treat behavioral problems. Furthermore, these generic issues inherent in all brain interventions are accompanied by other problems peculiar to specific applications. Examples are numerous, and their settings are various. Drugs used to treat attentiondeficit hyperactivity disorder (ADHD) are widely thought to be overused in children and may be surprisingly risky, even when indicated, in adults (El-Zein et al., 2005; Herman-Stahl, Krebs, Kroutil, & Heller, 2006). Despite their ubiquity, or perhaps because of it, drugs used to treat mood disorders are highly controversial especially when prescribed to children (President’s Council on Bioethics, 2003). Likewise, the profusion of imaging techniques for research on human behavior has recast a perennial clinical epistemological problem: the incidental finding linked in published research to a behavioral tendency out of character, literally, with the individual under examination (Illes & Chin, 2008; Miller, Mello, & Joffe, 2008).
214
ROBERT H. BLANK
POLITICAL IMPLICATIONS OF THE BRAIN SCIENCES The policy implications of the new neuroscience, then, are expansive and touch upon most areas of the human existence. Because of the availability of sensitive knowledge about the brains of individuals that accompanies diffusion of the applications outlined here, concern will center on what type of information should be collected, what to do with it once we have it, and who should have access to it. Because of the issues inherent in brain intervention (consent, stigmatization, privacy, autonomy, and so forth), these techniques also should be scrutinized as to their impact on the basic democratic concepts of freedom and equality. ‘‘Researchers are delving into the most intimate details of who we are, including such things as our personality traits, moral reasoning and tendency to violence. While their aims are commendable, many of their results raise big questions’’ (Editorial, 2004, p. 3). Similarly, Illes and Racine (2005) conclude that the ‘‘interpretation of neuroimaging data is a key epistemological and ethical challenge’’ and to that I would add a ‘‘policy challenge’’ as well. Complicating the problem of how we handle our new knowledge of the brain is a tendency to exaggerate the significance of this research and ignore its limitations (Arnason, 2010). In most imaging studies, even the statistically significant differences in brain structure and function are often very subtle and limited, but extrapolation is alluring and commonplace in media headlines because its subject (the human brain and behavior) is so important and innately dramatic. However, as Garland suitably warns: the use of flawed or incomplete science, or the reliance on scientific predictions beyond what the science is prepared to support, are exactly the kinds of concerns that should be foremost in the public mind when contemplating the potential impact of predictive technologies or techniques. (2004, p. 1)
Roskies makes important distinction by delineating two main divisions of neuroethics: the ethics of neuroscience and the neuroscience of ethics (Roskies, 2002). In turn, the ethics of neuroscience comprises the ethical issues raised in the course of designing and executing neuroscientific studies (the ethics of practice) and the evaluation of the ethical and social impact that the results of those studies might have on existing social, ethical, and legal structures (the ethical implications of neuroscience). The second major category for Roskies is the neuroscience of ethics. Traditional ethical theory has centered on philosophical notions such as free will, self-control, personal identity, and intention that can now can be investigated from the perspective
Brain Sciences and Politics: Some Linkages
215
of brain function through imaging studies (H€ayry, 2010). While the ethics of neuroscience has clear political and public policy implications, the neuroscience of ethics is likely to remain the domain of philosophers and academicians for some time to come (Farah, 2005; Illes & Bird, 2006). Attention here, therefore, is focused on the former category where neuroethics most evidently intersects with public policy.
The Brain and Addictions One set of issues centers on addictive behaviors, which extract a huge social cost. Addiction affects all organs but it mostly affects the brain. Studies utilizing PET scans and MRIs with addicts in craving or withdrawal have proliferated (Hall, Carter, & Morley, 2004). A major finding is that all abused substances, no matter what the mechanism, stimulate the brain’s reward system and induce feelings of pleasure that can override even the basic survival activities. This explains why addicts are willing to substitute one substance for another. When exposed to these addictive substances, tolerance builds up within certain neurotransmitters. If the brain is excessively stimulated over time, it learns to adjust and its functions change. The brain is then affected directly. The two most important tracks of current research are the neural and genetic bases of addiction and the impact of addictive substances on the brain and its functioning (see Buchman, Skinner, & Illes, 2010). While this research is moving ahead very rapidly, there is little corresponding progress in the area of social interpretation or analysis of the findings that pose questions over antidrug policy, the legalization of marijuana, alcohol policy, and other issues of addiction. Recently, analogous research has been initiated on eating disorders (Vocks et al., 2010) and obesity (Taylor, Curtis, & Davis, 2009).
Sex Differences in the Brain Another set of potentially explosive political issues flows from the findings of research on the human brain that reveal structural and functional differences by sex. Even citing the findings of these studies is judged by some to be politically incorrect. However, many studies in the brain sciences have demonstrated striking variation between the brains of males and females, both on average and in normal distributions. One of the targets of such research is the hypothalamus (the regulatory center of primal activities) and
216
ROBERT H. BLANK
its 12 nuclei. Some of the nuclei are larger in males than they are in females; some are spherical in males and elongated in females. There are also differences in metabolic activity: men have higher levels of activity in the temporal limbic region and women have higher levels of activity in the posterior cingulate gyrus region. The corpus callosum, which connects the two hemispheres, is much larger in women. One might ask why it matters if there are differences in the brain across the sexes? It matters in terms of any policies we establish, particularly in developing and prescribing drugs that might operate differently on brains in men and women. Another serious political issue centers on education. According to the brain research, it might be counterproductive to treat boys and girls the same when it comes to education and social training. For example, some school systems have eliminated recess even though boys might be more prone to need physical breaks because their brains might need that time off.
The Brain, Emotions, and Aggression New knowledge about the human brain also challenges widely shared assumptions about altruism, love, lying, truth telling, personality, sexual orientation, decision-making, political ideology, racial prejudice, moral reasoning, and free will itself. Among the most significant political findings are those bearing upon aggression (Wahlund & Kristiansson, 2009; Eisenberger, Way, Taylor, Welch, & Lieberman, 2007). Special attention has focused on certain neurotransmitters, hormones, the limbic system, and the ‘‘fear-circuit’’ in the prefrontal cortex (Coccaro, McCloskey, Fitzgerald, & Phan, 2007). Moreover, aggression is studied under a constellation of affective antecedents and behavioral parallels: fear, anger, hatred, antisociality, impulsivity, and violence. Although specific genetic or neuronal patterns have not yet been found that explain aggression in humans generally, certain anomalies clearly predispose a person to impulsivity and paroxysmal aggression (Raine, 2008; Glenn & Raine, 2009; National Institute on Drug Abuse, 2010). Aggression can be made more likely when the prefrontal cortex, among other higher brain regions, does not adequately moderate impulses originating in the limbic system. Moreover, lesions in the prefrontal cortex or disrupted functioning of the neurotransmitter serotonin can also impair inhibitory fear reduction, thus leading to aggression (Mobbs, Lau, Jones, & Frith, 2007). As a result, contemporary research has focused on the inhibitory role of serotonin first noticed when studies found that people who
Brain Sciences and Politics: Some Linkages
217
became aggressive under the influence of alcohol made less serotonin than those who did not become so. Since then, abnormalities of serotonin metabolism (its production, storage, release, or reuptake) have been found disproportionately in impulsively aggressive men, compulsive fire-setters, and the violently suicidal. Various studies have found a positive association between lower serotonin concentrations and an inclination to be aggressive. In addition, dopamine, norepinephrine, epinephrine, gamma aminobutyric acid (GABA), glycine, glutamic acid, and endorphins have all attracted investigation. Moreover, imaging data have shown notably increased prefrontal activity in subjects exposed to violent images, a finding consistent with the filtering hypothesis and suggesting that aggressive disorders might be explained by a ‘‘shortcircuiting’’ in this region. Early developmental damage here has been associated with impairment of social and moral reasoning, as is common in psychopaths (Schmook et al., 2010). Brain scans of convicted murderers have found drastically decreased prefrontal activity, particularly among those who had killed on impulse, rather than following premeditation. Compared to controls, murderers also showed increased activity in the amygdalas and decreased activity in the corpus callosum, the latter finding suggesting deficient interhemispheric communication. The causes of such dysfunctions are not well understood, although birth injury, childhood infection, neurotoxicant exposure, trauma, and violence itself are obvious candidates. Teicher (2002), for instance, found that the negative effects of child abuse on the functioning and development of the brain are long term and enduring. The impact of child abuse is not simply behavioral; it actually changes the structure and functioning of the brain or of particular neurotransmitters and their receptor sites. Functional studies of neuroanatomy often now focus on the amygdala, a collection of nuclei in an almond-shaped area in the medial temporal lobe, right and left. The amygdala is principally involved with emotion (ShinnickGallagher, 2003) and has been identified as key to the specific brain system that processes fear. As the major site of fear conditioning, the amygdalas receive danger signals and relay them in a manner that results in protective responses without conscious awareness. Not surprisingly, then, amygdala activity has been tied to generalized social phobia, a condition characterized by avoidance of situations where scrutiny by others is possible (Stein, Goldin, Sareen, Eyler Zorrilla, & Brown, 2002). The amygdala is also thought to play an important role in the neural circuitry of social intelligence and in the processing of social information (Adolphs, 2003). Furthermore, findings of numerous studies using a variety of imaging
218
ROBERT H. BLANK
techniques have highlighted the importance of the amygdala and other limbic regions in the identification of and response to emotionally salient information in the environment (Phillips, 2004). Specific abnormalities in identification and response to emotional material are associated with a variety of psychiatric disorders, and structural and functional abnormalities within these neural regions are commonplace in these populations. However, the amygdala deals not only in fear but also in happiness, sadness, unpleasantness, emotional memory, stimulus-reward learning, and prejudice (Coccaro et al., 2007; Baxter & Murray, 2002; Blair, 2003; Morrison, Allardyce, & McKane, 2002). Insights here are many for an understanding not only aggression but also discipline, corrections, addictions, mood disorder, psychosis, xenophobia, panic, and so on.
Free Will and Individual Responsibility Ultimately the impact of the brain sciences comes down to the question of free will and individual responsibility (Buller, 2010). Does this new knowledge that much of our behavior is shaped by neurotransmitters and receptors or that we can design new drugs or interventions to alter the brain mean that individuals are no longer responsible for their own actions? Should we no longer assume that we have free will? Free will, or some concept like free will, is at the center of our retributive justice system. Moreover, democracy itself assumes that individuals have the capacity to make rational choices – not in the amygdala but in the neocortex, where citizens can think through the issues rationally and then make a choice. Clearly, the traditional view that voting is largely a rational choice is severely compromised by emerging knowledge about the brain that raises broad, theoretical, and conceptual questions concerning free will and individual responsibility (Custers & Aarts, 2010). One ensuing question that affects our changed understanding of moral reasoning and judgment, and thus criminal intention and responsibility, is the degree to which brain abnormalities should affect sentencing of law breakers (Arnason, 2010). Although brain damage, hormonal, and neurotransmitter arguments have been presented as defenses against conviction for violent crimes, to date rarely have such attempts been successful. In part, this is because the notion of brain damage or abnormality remains subjective, and any links to any specific behavior are now tenuous at best. However, while we are far from a full understanding how the brain influences behavior, and it is unlikely that we will ever be able to conclusively demonstrate that a
Brain Sciences and Politics: Some Linkages
219
behavior is caused by a single biochemical or environmental defect, the fact that such information is available will increase societal pressures to use it. Under these conditions, should society reduce the level of culpability or legal responsibility for criminals who are found to have abnormal brain scans? Although at least one court has rejected fMRI lie detection based on such knowledge (Miller, 2010), as noted by Dunagan, ‘‘neural imaging technologies stand to be a very controversial – and radical – new development in law and law enforcement’’ (2010, p. 1). At the minimum, contemporary research on the brain and free will and aggression will force wider debate of this issue. Furthermore, it will require reevaluation of the potential for coerced use of highly selective antiaggression drugs or implant chips in the prefrontal cortex to compensate for reduced brain function (Enserink, 2000). A question already facing the courts is whether to order medication for mentally ill defendants so that they can stand trial. In the near future, this debate is likely to shift to the use of preemptive mandated interventions to modify the brain chemistry of those persons diagnosed through brain imaging tests as being at risk for violent behavior. Paradoxically, new knowledge about the interaction of the brain and the childhood environment also intensifies the need to develop preventive policies in response to early maladoptive behaviors of males in particular.
The Brain and Intergroup Conflict In a related area, there is considerable research on the neural foundations of bias and its ties to intergroup conflict. Imaging studies have found that even among seemingly unprejudiced people, racial category labels prime stereotypes (Fiske, 2002). Out-group cues such as faces or names activate negative evaluative terms. Brain imaging studies consistently demonstrate an activation of the amygdala in response to out-group faces, indicating fear or anxiety (Phelps, 2001). Importantly, these biases appear automatic, unconscious, and unintentional, thus creating a subtly hostile environment to out-groups that eludes rational choice. According to Phelps et al., the ‘‘correlation between amygdala activation and indirect measures of race bias suggests that the amygdala is engaging in the automatic processing of social group information from facial stimuli’’ (2003, p. 204). A fundamental question is what triggers the amygdala to preset these distinctions in the first place, presumably at a very early age. These data not only have important implications for domestic racial harmony but also are critical in studying
220
ROBERT H. BLANK
international conflict and terrorism – how our brains identify those who are a threat, for example, the enemy (Long & Brecke, 2003). As Pinker (2002) notes, diplomats often try to bring the opposing sides together by facilitating trade, cultural exchanges, and people-to-people exchanges. They try to blunt long-standing animosities by fashioning compromises that allow each side to save face and bring in third parties as guarantors, but are often frustrated when at the end of the day the two sides seem to hate each other as much as they did before the intervention. Research on the role of the limbic system suggests that dependence on rational incentives is unlikely in itself to resolve the fears and hatreds and that even the leaders of the opposing sides can be limited in their ability to convince their followers to accept peace without ameliorating emotional biases deep within the brain. These brain-based mechanisms shaped by fearful environments might help explain why long-standing problems in the Middle East, Northern Ireland, Somalia, Cambodia, the Balkans, and elsewhere appear irresolvable no matter how much effort is expended. How can policy makers best integrate this emerging knowledge of the brain’s influence on aggression, bias, and intergroup conflict into domestic and foreign policy? Are there more effective ways to deal with such problems than force or traditional mediation with handpicked leaders? What good does it do to remove a ruthless dictator if his followers hold similar views of outsiders? Contrarily, what dangers are there when leaders identify certain easily identifiable ethnic groups as the enemy? Or when they profile based on physical characteristics? What deep-seated evolutionary fear circuits do such policies trigger? To what extent do leaders exploit the fear system to build support for their policies? It is probable that we will never fully understand, much less be able to predict or treat, aggression by looking only at the genes and brain because at their base they are political and social problems, not just biological ones. However, the phenomena we call social and political, themselves, represent shared understandings among individuals, not external happenings that ‘‘mysteriously affect’’ human affairs. So, although a better knowledge of the role of the brain and human behavior is not sufficient to explain human aggression, ‘‘one cannot understand violence without a thorough understanding of the human mind’’ (Pinker, 2002).
Commercial Uses of the Brain Sciences A final set of concerns centers on the prospective commercial uses of neuroscience. Relatively innocuous applications of brain imaging such as the
Brain Sciences and Politics: Some Linkages
221
rapidly growing area of neuromarketing strike many observers as highly problematic (Fisher, Chin, & Klitzman, 2010; Singer, 2004). Although it’s hard to get an accurate accounting because many companies don’t publicize such activities, neuromarketing research in academic institutions and neuromarketing consulting start-ups around the world are escalating (Dunagan, 2010). Critics see neuromarketing as dangerous because marketers can bypass our neural control systems, or hijack our free will. Reality of course is more complicated concerning direct-to-brain marketing since seeing the expression of desire in brain readings does not always translate to predictable purchasing behavior. The brain–behavior link is exceedingly complex, thus even with powerful new tools to image and find neural correlates of desire, neuromarketing may be useful, but it is no more terrifying, or terrifyingly effective, than a good focus group according to Dunagan (2010). Other even more contentious commercial applications such as lie detection (Wolpe, Foster, & Langleben, 2005) are already proliferating, and there most likely will be pressures to utilize brain scans as a screening device for job and school applicants and other nonmedical uses in the future. Phillips is particularly concerned with the potential use of brain scans to diagnose or make predictions about a behavior or illness: ‘‘who might be a paedophile (sic), who might be violent, who might develop a mental illness or prion disease y ?’’ (2004). As with our unfolding knowledge of the human genome, our burgeoning knowledge of the brain can be a double-edged sword. Imaging genetics, especially, has considerable potential to become an attractive product in emerging high-technology medical markets. Guidance for considering the power of the combined technologies may well be learned from past experiences with commercialization. These include considerations of the sale of technology for unintended uses and for which data do not exist, sustained validity, and the impact of aggressive marketing strategies that pose risks in particular to some of the most vulnerable members of society who suffer from neurological or mental health disorders and cognitive deficits (Illes et al., 2004). Neuroscience databases, including those archiving neurological and genetic data, are being created throughout the world. The resulting challenges for the commercial sector, as for researchers and policy makers, are enormous. A range of legal authorities and potential litigants, such as insurers and employers, may seek this information. Likewise, individuals who are data sources may also try to access information about themselves. As the data can be banked indefinitely, these issues have an indeterminate lifetime for a source individual and family members.
222
ROBERT H. BLANK
NEUROSCIENCE POLICY The move of neuroscience into the policy domain alters the context by bringing to the fore political considerations and divisions and placing the resolution of these issues in the milieu of interest group politics. Given the vast economic, social, and personal stakes surrounding the brain sciences, this is unavoidable. Moreover, because neuroscience is laden with ethical overtones that defy easy resolution, achieving consensus and building policy will be problematic. The issues raised by the brain sciences are evidenced in the almost daily announcements of new research findings and in the development of innovative technological applications. Intervention in the brain is an especially controversial policy area because of the rapid succession of advances in knowledge and the shortened lag time between basic research and the diffusion of an array of applications. As noted by Hyman, the potential use of brain imaging ‘‘to reconstruct a person’s recent experience or investigate his or her veracity’’ not only raises traditional bioethical questions of privacy but also engages broader communities who are not often represented in discussions of bioethics (Hyman, 2007). Although many of the specific issues raised by neuroscience applications are unique, the policy dimensions are essentially similar to other areas of biomedical research. At their base, there are three policy dimensions relevant to neuroscience. First, decisions must be made concerning the research and development of the technologies. Because a considerable amount of research in the brain sciences is funded either directly or indirectly with public funds, it is important that public input be included at this stage. The heightened prominence given technological forecasting and assessment of the social as well as technical consequences of biomedical technologies early in the research and development phase represents one means of incorporating broader public interests. One problem is how to best design assessment processes so as to evaluate the social/political impact of neuroscience in addition to efficacy and long- and short-term safety. The need for anticipatory policy making is crucial but, to date, largely absent. The second policy dimension relates to the individual use of technologies once they are available. Although direct government intrusion into individual choice in the health arena has, until recently, been limited, the government does have at its disposal an array of more or less explicit devices to encourage or discourage individual use. They include tax incentives, provision of free or subsidized services, and education programs. Although conventional regulatory mechanisms may generally be effective in protecting potential users or targets of new neuroscience applications, it is critical that
Brain Sciences and Politics: Some Linkages
223
their effectiveness be assessed and monitored. Neuroscience policy, like genetic policy, has distinctive importance in contemporary politics because it challenges intensely held societal values relating to the self, privacy, discovery, justice, health, and rights. The third dimension of biomedical policy centers on the aggregate societal consequences of widespread propagation of a technology. For instance, how might the use of neural imaging to type personalities affect our concept of equality of opportunity or of respect among individuals? How might neural (and genetic) imaging techniques be used in employment and for insurance and what social and economic impacts will this have? What impact will the diffusion of neural grafting have on the provision of health care? To what extent should we allow, or even encourage, the use of psychotropic drugs on children? What about nootropics? Policy making here requires a clear conception of goals, extensive data to predict the consequences of each possible course of action, an accurate means of monitoring these consequences, and mechanisms to cope with the consequences if they are deemed undesirable. Moreover, the government has a responsibility to ensure quality control standards and fair marketing practices. The governmental response to developments in the brain sciences can take many forms and can occur any time from the earliest stages of research to the provision of specific techniques. Brain policy like any other policy, then, can be permissive, affirmative, regulatory, or prohibitive. The government always has the option of taking no action, thus permitting any actions by the private sector. It can also make affirmative policies that promote or encourage certain activities, for example, government funding of research or provision of services to facilitate more widespread use of a particular technique or procedure. The question of whether the government ought to be providing such encouragement, and if so by what means, will likely be a subject of political debate. Should public funds be used to pay for expensive interventions for patients who cannot afford them? Should private insurers be required by law to cover the potentially unlimited expense of treating an Alzheimer’s patient? In some instances, the line between encouraging and coercing or mandating use is a fine one. Although regulatory policy can be framed to apply only to governmentsupported activities, usually it consists of rules governing activities in both the public and private sectors. Regulation can be used to ensure that standards of safety, efficacy, and liability are adhered to, and unlike professional association guidelines, regulations have the force of law and often include legal sanctions for violations. Lastly, although far less common than regulation, prohibitive policies could be implemented that impose legal
224
ROBERT H. BLANK
sanctions on a specific research activity or application such as has been done with human cloning. Another type of prohibitive policy is to preclude public funding of specific areas of research and development (e.g., certain types of human fetal, embryo, or stem cell research) or specific procedures such as ECT. It remains to be seen what, if any, areas of brain intervention or research are candidates for prohibition, but any such policy will reflect political motives or be a response to demands of interest groups, often on moral grounds. Fig. 1 illustrates the range of policy options available. Many of these options have been used in various political jurisdictions with regard to stem
Favor Intervention Mandate use
Support complete individual choice
Fund public research Favor free market -commercialization with no government intervention
Incentives for private research Encourage individual use -tax incentives -education -free services
Professional guidelines Private policy
Consumer protection Set standards of practice
Bioethical deliberation
Favor government involvement
Oppose government involvement
Monitor social consequences No public funding Critical assessment Regulate marketing practices
Fear mandates, social control, stigmatization, Big Brother
Discourage individual use Strict regulation Prohibit use
Oppose Intervention
Fig. 1.
The Role of Government in the Brain Sciences.
Brain Sciences and Politics: Some Linkages
225
cell research, reproductive technologies, genetic technologies, or earlier brain intervention techniques. They clearly demonstrate the wide divergence of policy responses available as well as the diametrically opposed positions on the role of the government that are found among various segments of society. Given the record in these related fields, there appears very little likelihood of anything approaching a consensus emerging either on the role of government in the brain sciences or the preferred policies regarding specific applications.
CONCLUSIONS As illuminated here, the brain sciences have made tremendous strides in the last two decades, in large part because of the inclusion of interconnected sciences and the availability of innovative technologies in imaging and computerization. However, as noted earlier, society has not kept pace to effectively frame the political issues flowing from these technologies. There is little evidence of anticipatory policy, or even timely reactive policy for that matter. In addition, the inclusion of the brain in a more interactive paradigm of human behavior threatens the prevailing and deeply entrenched social science model that resists the findings of brain research pertaining to human behavior. Finally, because neuroscience challenges fundamental beliefs on both ends of the ideological spectrum, those on both the political left and the right will attempt to interpret the findings in the most favorable light to serve their purposes and reject findings that contradict their views. In all of these areas – public policy, social science, and politics – the brain sciences promise to become a most contentious and divisive force. Although the brain sciences have not yet received the attention of other areas of human science, its implications warrant close scrutiny by political scientists. One must be skeptical of wild speculation on all sides, but too much is at stake to ignore the findings of neuroscience or to leave to chance how they are interpreted and applied. No one knows how far or in what direction the brain sciences will take us, but we can be sure that as the cutting edge for medicine as well as human behavior, the brain sciences, in combination with analogous advances in genetics, molecular biology, and most likely nanoscience, will radically change our views of the human condition.
226
ROBERT H. BLANK
NOTE 1. Although there are nuances, the terms brain sciences and neuroscience are used interchangeably in this chapter.
REFERENCES Adolphs, R. (2003). Is the human amygdala specialized for processing social information? Annals of the New York Academy of Sciences, 985, 326–340. Arnason, G. (2010). Neuroimaging, uncertainty, and the problem of dispositions. Cambridge Quarterly of Healthcare Ethics, 19, 188–195. Baxter, M. G., & Murray, E. A. (2002). The amygdala and reward. Nature Neuroscience, 3, 563–574. Beasley, D. (2003). Researchers compile ‘atlas’ of the brain. Reuters News. www. news.yahoo.com/news?tmpl¼story2&cid¼570&u¼/nm/20030808/sc_nm/health_brain.html Bell, E., Mathieu, G., & Racine, E. (2009). Preparing the ethical future of deep brain stimulation. Surgical Neurology, 72, 577–586. Blair, R. J. R. (2003). Neurobiological basis of psychopathy. The British Journal of Psychiatry, 182, 5–19. Blank, R. H. (1999). Brain policy: How the new neuroscience will change our lives and our politics. Washington, DC: Georgetown University Press. Blank, R. H. (2006). The brain, aggression and public policy. Politics and the Life Sciences, 24(1-2), 12–21. Brain Fingerprinting Laboratories. (2003). Scientific procedure, research, and applications. Available at www.brainwavescience.com/technologyoverview.php. Buchman, D. Z., Skinner, W., & Illes, J. (2010). Negotiating the relationship between addiction, ethics, and brain science. AJOB Neuroscience, 1(1), 36–45. Buller, T. (2010). Rationality, responsibility and brain function. Cambridge Quarterly of Healthcare Ethics, 19, 196–204. Carney, S., & Geddes, J. (2003). Electroconvulsive therapy. British Medical Journal, 326, 43. Chance, B., Anday, E., Nioka, S., et al. (1998). A novel method for fast imaging of brain function, non-invasively with light. Optics Express, 1998(2), 411. Coccaro, E. F., McCloskey, M. S., Fitzgerald, D. A., & Phan, K. L. (2007). Amygdala and orbitofrontal reactivity to social threat in individuals with impulsive aggression. Biological Psychiatry, 62, 168–178. Custers, R., & Aarts, H. (2010). The unconscious will: How the pursuit of goals operates outside of conscious awareness. Science, 329(5987), 47–50. Dillow, C. (2010). New optogenetic neural implants use precise beams of light to manipulate the brain. Available at www.popsci.com/science/article/2010-02 Dunagan, J. (2010). Six neurocentric industries. The institute for the future. Available at http:// www.good.is/post/six-neurocentric-industries/. Editorial. (2004). Misplaced faith in ‘mind reading’ scans is sure route to injustice. New Scientist, July 31, p. 3. Eisenberger, N. I., Way, B. M., Taylor, S. E., Welch, W. T., & Lieberman, M. D. (2007). Understanding genetic risk for aggression: Clues from the brain’s response to social exclusion. Biological Psychiatry, 61, 1100–1108.
Brain Sciences and Politics: Some Linkages
227
El-Zein, R. A., Abdel-Rahmanb, S. Z., Hayb, M. J., Lopeza, M. S., Bondya, M. L., Morris, D. L., & Legator, M. S. (2005). Cytogenetic effects in children treated with methylphenidate. Cancer Letters, 230(2), 284–291. Enserink, M. (2000). Searching for the mark of Cain. Science, 289, 575–579. Farah, M. J. (2005). Neuroethics: The practical and the philosophical. Trends in Cognitive Sciences, 9(1), 34–40. Farah, M. J. (Ed.) (2010). Neuroethics: An introduction with readings. Cambridge, MA: MIT Press. Fisher, C. E., Chin, L., & Klitzman, R. (2010). Defining neuromarketing: Practices and professional challenges. Harvard Review of Psychiatry, 18(4), 230–237. Fiske, S. T. (2002). What we know about bias and intergroup conflict: The problem of the century. Current Directions in Psychological Science, 11(4), 123–128. Garland, B. (2004). Neuroscience and the law. Professional Ethics Report, 17(1), 1–3. Gazzaniga, M. S. (2005). The ethical brain. New York: Dana Press. Giordano, J. J., & Gordijn, B. (Eds). (2010). Scientific and philosophical perspectives in neuroethics. Cambridge, UK: Cambridge University Press. Glenn, A. L., & Raine, A. (2009). Psychopathy and instrumental aggression: Evolutionary, neurobiological, and legal perspectives. International Journal of Law and Psychiatry, 32, 253–258. Hall, S. S. (2003). The quest for a smart pill. Scientific American, 289(3), 54–65. Hall, W., Carter, L., & Morley, K. I. (2004). Neuroscience research on the addictions: A prospectus for future ethical and policy analysis. Addictive Behaviors, 29, 1481–1495. H€ayry, M. (2010). Neuroethical theories. Cambridge Quarterly of Healthcare Ethics, 19, 165–178. Herman-Stahl, M. A., Krebs, C. P., Kroutil, L. A., & Heller, D. C. (2006). Risk and protective factors for methamphetamine use and nonmedical use of prescription stimulants among young adults aged 18 to 25. Addictive Behaviors, 32(5), 1003–1015. Hurley, R. A., & Taber, K. H. (2008). Windows to the brain: Insights from neuroimaging. New York: American Psychiatric Publishing, Inc. Hyman, S. E. (2007). Neuroethics: At age 5, the field continues to evolve. The Dana Foundation, www.dana.org. Illes, J., & Bird, S. J. (2006). Neuroethics: A modern context for ethics in neuroscience. Trends in Neurosciences, 30(10), 1–7. Illes, J., & Chin, V. N. (2008). Bridging philosophical and practical implications of incidental findings in brain research. The Journal of Law, Medicine & Ethics, 36(2), 298–301. Illes, J., & Racine, E. (2005). Imaging or imagining? A neuroethics challenge informed by genetics. American Journal of Bioethics, 5(2), 5–18. Illes, J., Rosen, A. C., Huang, L., et al. (2004). Ethical consideration of incidental findings on adult brain MRI in research. Neurology, 62, 888–890. Kempf, L., & Meyer-Lindenberg, A. (2006). Imaging genetics and psychiatry. Focus, 4(3), 327–338. Long, W. J., & Brecke, P. (2003). The emotive causes of recurrent international conflicts. Politics and the Life Sciences, 22(1), 24–35. Meyer-Lindenberg, A., Nicodemus, K. K., Egan, M. F., Callicott, J. H., Mattay, V., & Weinberger, D. R. (2008). False positives in imaging genetics. NeuroImage, 40, 655–661. Miller, F. G., Mello, M. M., & Joffe, S. (2008). Incidental findings in human subjects research: What do investigators owe research participants? The Journal of Law, Medicine & Ethics, 36(2), 271.
228
ROBERT H. BLANK
Miller, G. (2010). fMRI lie detection fails a legal test. Science, 328(5984), 689–690. Mobbs, D., Lau, H. C., Jones, O. D., & Frith, C. D. (2007). Law, responsibility, and the brain. PLOS Biology, 5, 693–700. Morrison, P. D., Allardyce, J., & McKane, J. P. (2002). Fear knot: Neurobiological disruption of long-term fear memory. The British Journal of Psychiatry, 180, 195–197. € Muller, S., & Walter, H. (2010). Reviewing autonomy: Implications of the neurosciences and the free will debate for the principle of respect for the patient’s autonomy. Cambridge Quarterly of Healthcare Ethics, 19, 206–217. National Institute on Drug Abuse. (2010). Impulsive-antisocial personality traits linked to a hypersensitive brain reward system. Mental Health Weekly Digest, May 2010, p. 31. Phelps, E. A. (2001). Faces and races in the brain. Nature Neuroscience, 4(8), 775–776. Phelps, E. A., Cannistraci, C. J., & Cunningham, W. A. (2003). Intact performance on an indirect measure of race bias following amygdala damage. Neuropsychologia, 41, 204–206. Phillips, H. (2004). Private thought, public property. New Scientist, July 31, pp. 38–41. Pinker, S. (2002). The blank slate: The modern denial of human nature. New York: Viking. President’s Council on Bioethics. (2003). Beyond therapy: Biotechnology and the pursuit of happiness. Washington, DC: U.S. Government Printing Office. Racine, E. (2010). Pragmatic neuroethics: Improving treatment and understanding of the mindbrain. Cambridge, MA: MIT Press. Raine, A. (2008). From genes to brain to antisocial behavior. Current Directions in Psychological Science, 17, 323–328. Reuters. (2005). Panel to review pacemaker device for brain. Available at ww.msnbc.msn.com/ id/5190146/ Rose, S. P. R. (2002). ‘Smart drugs’: Do they work? Are they ethical? Will they be legal? Nature Reviews/Neuroscience, 3, 975–977. Roskies, A. (2002). Neuroethics for the new millennium, commentary. Neuron, 35, 21–23. Schlaepfer, T. E., Lisanby, S. H., & Pallanti, S. (2010). Separating hope from hype: Some ethical implications of the development of deep brain stimulation in psychiatric research and treatment. CNS Spectrums, 15(5), 285–287. Schmetz, M. K., & Heinemann, T. (2010). Ethical aspects of deep brain stimulation in the treatment of psychiatric disorders. Fortschritte der Neurologie-Psychiatrie, 78(5), 269–278. Schmook, M. T., Brugger, P. C., Weber, M., Kasprian, G., et al. (2010). Forebrain development in fetal MRI: Evaluation of anatomical landmarks before gestational week 27. Neuroradiology, 52(6), 495–499. Shinnick-Gallagher, P., Pitkanen, A., Shekhar, A., Cahill, L.,(Ed.) (2003). The amygdala in brain function: Basic and clinical approaches. New York: Academy of Sciences. Singer, E. (2004). They know what you want. New Scientist, July 31, pp. 36–37. Stein, M. B., Goldin, P. R., Sareen, J., Eyler Zorrilla, L. T., & Brown, G. G. (2002). Increased amygdala activation to angry and contemptuous faces in generalized social phobia. Archives of General Psychiatry, 59(11), 1027–1034. Taylor, V. H., Curtis, C. M., & Davis, C. (2009). The obesity epidemic: The role of addiction. CMAJ, 182(4), 327–328. Tairyan, K., & Illes, J. (2009). Imaging genetics and the power of combined technologies: A perspective from bioethics. Neuroscience, 164(1), 7–15. Teicher, M. H. (2002). Scars that won’t heal: The neurobiology of child abuse. Scientific American, 286(3), 68–75.
Brain Sciences and Politics: Some Linkages
229
Viding, E., Williamson, D. E., & Hariri, A. R. (2006). Developmental imaging genetics: Challenges and promises for translational research. Development and Psychopathology, 18, 877–892. € Vocks, S., Busch, M., Gronemeyer, D., Schulte, D., Herpertz, S., & Suchan, B. (2010). Neural correlates of viewing photographs of one’s own body and another woman’s body in anorexia and bulimia nervosa: An fMRI study. Journal of Psychiatry and Neuroscience, 35(3), 163–176. Wahlund, K., & Kristiansson, M. (2009). Aggression, psychopathy and brain imaging – Review and future recommendations. International Journal of Law and Psychiatry, 32, 266–271. Wolpe, P. R., Foster, K. R., & Langleben, D. D. (2005). Emerging neurotechnologies for liedetection: Promises and perils. American Journal of Bioethics, 5(2), 39–49.
BRAIN IMAGING AND POLITICAL BEHAVIOR: A SURVEY John M. Friend and Bradley A. Thayer INTRODUCTION: THE POLITICAL BRAIN Political science is often derided for being a ‘‘soft’’ science, one unable to generate hard predictions about political behavior, or without the ability to test its hypotheses, unlike physics, biology, or, among the social sciences, economics. Standards of hypothesis testing, data collection, and testing were unfairly seen to be lacking in comparison with the hard sciences. Accordingly, political scientists often had to struggle to have the knowledge produced about political behavior taken seriously. It would not be too remiss to identify an inferiority complex among political scientists, when they discussed the pantheon of scientific disciplines and their low position in it. Yet, the advance of science, specifically, neuroscience and cognitive psychology, holds great promise for political science. Their advance means that the days of the scientific inferiority complex are at an end. Remarkably, this change is too often unrecognized by political scientists themselves. The fundamental objective of this chapter is to call the attention of our colleagues in political science to the profound changes forever altering our discipline. In 2007, in an op-ed for the New York Times, Marco Iacoboni and colleagues claimed that their recent study of a group of swing voters revealed ‘‘some voter impressions on which the 2008 election may well turn’’
Biology and Politics: The Cutting Edge Research in Biopolitics, Volume 9, 231–255 Copyright r 2011 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 2042-9940/doi:10.1108/S2042-9940(2011)0000009012
231
232
JOHN M. FRIEND AND BRADLEY A. THAYER
(Iacoboni et al., 2007). Studies of voting behavior and public opinion positing insight into election results are nothing new, but what made this particular study unique (and highly controversial) was its use of functional magnetic resonance imaging (commonly known as fMRI) to interpret and possibly predict voting behavior in the 2008 presidential election. Through an observation of the brain activity of 20 registered swing voters as they answered questions about their political preferences, the study documented what regions of the brain responded to pictures and videos of presidential candidates. Interestingly though, the mainstream response to the study was sharply divided. For some, the use of brain imaging to better understand voting behavior and partisan preference is seen as a promising new approach to the study of politics (Tingley, 2006), while, for others, the reliance on fMRI to investigate political behavior is considered to be a ‘‘mind game’’ or a ‘‘mindless obsession’’ at best (Anon, 2007; Barber, 2008). Decades before the proliferation of brain imaging studies addressing the neurocognitive correlates of political orientation, partisanship, and voting behavior, the social brain hypothesis of evolutionary psychology provided an understanding of human social nature that has recently intersected with breakthroughs in neuroscience and behavioral genetics to produce a wealth of new information about political behavior. Before we can begin to appreciate the contributions of fMRI and positron emission tomography (PET), along with other brain imaging techniques, to theories of the mind that hold grave insights into intergroup conflict and cooperation, decisionmaking, and prejudice, the social psychological mechanisms (and their adaptive functions) of the brain need be addressed. Essentially, as JeanPierre Changeux once put it, ‘‘the neuronal content of the black box can no longer be ignored’’ (Changeux, 1997, p. 97). Since the ‘‘Decade of the Brain,’’ numerous competing theories and perspectives have emerged seeking to unravel the relationship between neuronal substrates and human behavioral traits. While many have voiced a concern over the use of brain imaging to study the brain’s role in human behavior, numerous studies have shown that neuroscience techniques can contribute greatly and uniquely to our understanding of the regions of the brain involved with the ‘‘performance of a well-designed task’’ by ‘‘providing a broad and detailed view of the processing architecture of cognitively engaged networks’’ (Raichle, 2003, p. 760). In particular, we find the deeper understanding of human social behavior offered by brain imaging studies to greatly contribute to the growing field of biopolitics. Evolutionary psychology proposes that human behavior can be understood, in part, ‘‘as the product of evolved psychological mechanisms that depend
Brain Imaging and Political Behavior
233
on internal and environmental input for their development, activation, and expression in manifest behavior’’ (Confer et al., 2010, p. 110). Like our most basic intuitive feelings, the ongoing revolution in the biological sciences has shown that even the most complex political behaviors, such as cooperation and warfare, may be interconnected with ‘‘genetically programmed predispositions’’ shaped by natural selection (Alford & Hibbing, 2004). To answer the questions and test the theories that biopolitical theory has raised, cognitive neuroscience, and more recently political psychologists, have relied upon brain imaging techniques to identify the neural circuits associated with specific functions and behaviors that have social and political implications. It is clear that the brain imaging techniques of today remain limited in their ability to reveal fully the correlation between brain activity and cognitive processes, but despite these ‘‘growing pains,’’ many in the fields of neuroscience, evolutionary psychology, and politics see brain imaging, such as fMRI, as a promising new tool for investigating and interpreting the brain–behavior relationship, as well as a valuable technique for providing an overall better understanding of how the brain is physically wired. The developments in brain imaging and cognitive neuroscience theory during the past quarter century, as John Cacioppo and his colleagues point out, ‘‘have increasingly made it possible to investigate the differential involvement of particular brain regions in normal and disordered thought in humans’’ (Cacioppo, Berntson, & Nusbaum, 2008, p. 62). In other words, advances in brain imaging have enabled ‘‘researchers to bring distinctly human attributes y under scientific scrutiny’’ (Miller, 2008, p. 1413; also see Tingley, 2006). Taking into account the strengths and weaknesses of brain imaging and the wealth of new information provided by neuroscience techniques for the study of human political behavior, this chapter surveys the major contributions brain imaging studies have offered to the field of biopolitics. Since it is beyond the scope of this chapter to address the literature on brain imaging studies of political behavior in its entirety, we instead focus on three major areas of research that have political significance: (1) studies of aggression and violence in human social behavior, (2) prejudice and cooperation within in-group/out-group relations, and (3) a neuronal approach to voting behavior and partisan preference. We conclude with a discussion of the current limits of brain imaging techniques and the future possibilities for investigating the biological underpinnings of human sociopolitical behavior. Despite the controversy surrounding the application of fMRI to an analysis of political behavior, Marcus Raichle is correct to point out that brain imaging, and social neuroscience in general, ‘‘is clearly a
234
JOHN M. FRIEND AND BRADLEY A. THAYER
work in progress, but one that offers the potential of extending the relationship of brain science to a much broader range of social sciences, with important implications for the integration of natural and social sciences, as so eloquently discussed by E. O. Wilson’’ (Raichle, 2003, pp. 762–763).
MAPPING AGGRESSION AND VIOLENCE Before the triumph of brain imaging techniques, Albert Somit was one of the first to acknowledge that a ‘‘biologically oriented’’ understanding of political behavior would require an investigation into the brain. For Somit, the major breakthroughs in the field of psychopharmacology during the 1950s, particularly the development of psychoactive drugs and their influence on affective behavior and process, revealed ‘‘a political being whose behavior must be understood in terms both of social conditions (i.e., learning) and of inherent, biologically-transmitted tendencies toward certain types of responses, given certain types of environmental stimuli’’ (Somit, 1968, p. 560). In other words, psychopharmacology’s alteration of the ‘‘physiological or biological functioning of the human body’’ provided further evidence for a ‘‘direct link between biology and behavior in man as well as in other forms of life’’ (Somit, 1968, p. 560). Like Somit, Roger Masters anticipated the need to investigate the ‘‘neuronal content’’ of the brain in order to better understand the effects of neurochemistry on personality and political behavior. Pioneering the field of neuropolitics, Masters wrote insightfully on the implications of the ‘‘neurotransmitter revolution’’ to the study of human social behavior, particularly the relationship between serotonergic (5-HT) function and aggressive behavior (Masters & McGuire, 1994). By questioning the biological relevance of the legal norms surrounding issues of impulse control, criminality, and culpability, the work of Masters and colleagues can be seen as a direct challenge to traditional legal and public policy paradigms. More recently, neuroscience techniques and approaches have caught up with biopolitical theory, and in turn have provided stronger empirical evidence of the link between biology and human behavior. For those interested in the causes of ethnic and international conflict, one of the most complex and challenging dimensions of political behavior continues to be coalitional warfare. In fact, the archaeological and historical records as well as ethnographic research conducted by anthropologists and biologists reveal a long history of conflict and warfare in human societies (Keeley, 1997; Daly & Wilson, 1988). According to an evolutionary perspective, the ubiquity of
Brain Imaging and Political Behavior
235
intergroup conflict across cultures and societies should be no surprise when we take into consideration that certain behavioral traits, such as aggression, violence, and xenophobia, that is to say the necessary ingredients for prolonged conflicts, are evolved functions of natural selection (Thayer, 2000, 2003; Rosen, 2005). While it is not completely incorrect to conclude that coalitional aggression is similar to the adaptive significance of individual aggression, John Tooby and Leda Cosmides point out that ‘‘war is not simply individual aggression writ large.’’ Coalitional aggression, instead, ‘‘evolved because it allowed participants in such coalitions to promote their fitness by gaining access to disputed reproduction enhancing resources that would otherwise be denied to them’’ (Tooby & Cosmides, 1988, 2010). With the recognition that ‘‘anger is part of the basic biology of the human species’’ (Sell, Tooby, & Cosmides, 2009, p. 15073), studies on aggression, and its relationship to altruism, have become major contributions to the research on human political behavior. Just as Francis Crick noted the importance of studying one’s brain to better understand the human condition (Crick, 1979), contemporary brain imaging techniques have illuminated an underlying neurobiology of aggression in human social behavior (Siever, 2008). However, while noninvasive brain imaging techniques have provided a deeper understanding of the neuronal contributions to aggressive behavior, that is to say how the brain serves as an intermediate between genotype and social behavior, acts of violence, as well as other dimensions of human behavior, must not be viewed as solely an innate biological response. Although recent brain imaging studies have shown us that violent acts have genetic predispositions, such as in the case of impulsive aggression, the causes of human aggression, nevertheless, are clearly multifactorial, and include a wide array of cultural, socioeconomic, political, medical, psychological, and historical factors that can cause acts of violence. On this point, Robert Blank argues that the necessary question to ask is: ‘‘how, not whether, biology and environment are related – how a complex of biological factors interacts with a complex of environmental factors producing a single isolated aggressive behavior or a patterned concatenation of aggressive behaviors’’ (Blank, 2006, p. 15). To answer such a question, Blank further adds, ‘‘the role of the brain and its influence must be clarified’’ (Blank, 1999, p. 83). Brain mapping studies of the neural underpinnings of aggressive behavior have traditionally taken two methodological approaches: clinical studies of patients with brain lesions that display overt aggressive behavior and brain
236
JOHN M. FRIEND AND BRADLEY A. THAYER
imaging studies of patients with brain abnormalities that commit various forms of violent and/or antisocial acts (Paus, 2005). More recently, research on the brain–behavior relationship has begun to incorporate the influence of environmental stimuli on genotype, such as childhood mistreatment (Caspi et al., 2002) and exposure to neurotoxins (Masters, 2001a, 2001b). In the case of the latter, advances in brain imaging make the complexity of gene– environment interactions a more visible process, and, through a biological approach, ‘‘cognitive neuroscience and the behavioral effect of neurotoxins make it possible to discover environmental factors that have a remarkably strong effect on rates of violent behavior’’ (Masters, 2001a, p. 353). Our understanding of aggression has come a long way since the ‘‘instinct theory of aggression’’ proposed by Sigmund Freud (1922) and Konrad Lorenz (1974). For example, ongoing primate studies reveal that violent behavior is not so much the result of a ‘‘death instinct’’ or a ‘‘fighting instinct in beast and man which is directed against members of the same species,’’ but instead a product of a complex relationship between biological mechanisms and the physical environment. Ned Kalin posits that insights from primate models, specifically those of Rhesus monkeys (Macaca mulatta), can be used to better understand the causes of human aggression (Kalin, 1999). In particular, these models have shown that ‘‘aggressive behavior is related to other emotional states and affective disorders’’ and can be classified into three different categories that involve different neural mechanisms: defensive and fear motivated, offensive and impulsive, and self-injuries (Kalin, 1999, p. 29). Through the use of brain mapping techniques, certain regions of the brain have been linked to these different types of aggressive and violent behaviors. Most notably, the prefrontal cortex (PFC) ‘‘has been recognized as an important regulator of social and aggressive behavior’’ and additional findings suggest that this region of the brain ‘‘functions as a critical filter between the violent images themselves and the decisions people make in choosing how to respond to them’’ (Mozes, 2001). Likewise, neuroimaging evidence from studies of patients with acquired lesions to the PFC has revealed that prefrontal damage can result in a syndrome similar to psychopathy (Anderson, Bechara, Damasio, Tranel, & Damasio, 1999). Individuals with early-onset PFC damage, the most famous being Phineas Gage, demonstrate impaired social behavior, insensitivity to future consequences, and an inability to respond to punishment contingencies and behavioral interventions. With the case of Gage, such antisocial behavior was the result of trauma to the prefrontal region, but more recent functional imaging studies have shown that drug abuse can also impair the
Brain Imaging and Political Behavior
237
orbitofrontal cortex and cause psychopathic-like behavior and impulsive decision-making (London, Ernst, Grant, Bonson, & Weinstein, 2000). Furthermore, Davidson and colleagues argue that impulsive aggression and violence can be associated with a dysfunction between the interactions of the core neural substrates, particularly the PFC, the amygdala, hippocampus, hypothalamus, anterior cingulated cortex (ACC), insular cortex, ventral striatum, and other interconnected structures involved in the ‘‘various aspects of emotion, affective style, and emotion regulation’’ (Davidson, Putnam, & Larson, 2000 p. 591). Among these, the amygdala has received the most attention due to its active role in the regulation of emotion, specifically fear, and its appraisal of danger during fight-or-flight scenarios (LeDoux, 1995; Arnsten, 1998). Impulsive, affective aggression is believed to occur when the PFC is incapable, usually due to lesions, of inhibiting the amygdala and suppressing negative emotions.
THE LINK BETWEEN SEROTONIN AND AGGRESSION At the neurotransmitter level, dysfunction in emotional regulation can produce a disruption in the serotonergic (5-HT) system, specifically a reduction of the cerebrospinal fluid (CSF) of 5-HT metabolite 5-hydroxyindoleacetic acid (5-HIAA), which has been shown to cause impulsive, aggressive, and violent behavior (Davidson et al., 2000, p. 592). Interestingly, studies of both children with disruptive behavior disorder and recidivist violent adults show that 5-HIAA levels can predict future aggressive behavior (Kruesi et al., 1992; Virkkunen, De Jong, Bartko, & Linnoila, 1993; Siever & Trestman, 1993). From an evolutionary perspective, the relationship between lower 5-HT levels and aggressive behavior makes sense. Serotonin is believed to regulate our most primitive drives and emotions, such as sexual desire, mood, appetite, sleep, pain, and aggression (Kotulak, 1997, p. 68). Aggressive behavior, like the rest of these basic human emotions, is not necessarily bad. As stated earlier, aggression is considered to be an adaptive function that allowed our ancestors to compete for resources, mates, and status. Even today, in a social environment much different from that of our ancestors, aggression, usually in the form of competition and self-confidence, allows us to overcome obstacles and reach our desired goals.
238
JOHN M. FRIEND AND BRADLEY A. THAYER
Problems occur, however, when the biological mechanisms associated with aggression are inappropriately triggered by cues and stimuli that once posed a threat to survival. A classic example is the fear of snakes even though snakes no longer represent a serious daily threat to our survival. For some individuals, on the other hand, aggressive and violent behavior is common, and in some instances uncontrollable, due to a genetic or acquired deficiency in the serotonergic system. For individuals with low 5-HT levels, impulsive behaviors, such as ‘‘out-of-place’’ aggressive responses, antisocial violent crime, and sexual abuse (Kotulak, 1997; Tiihonen et al., 2001) are an inescapable reality that requires medical attention and assistance.
THE EFFECTS OF TESTOSTERONE ON AGGRESSIVE BEHAVIOR Like serotonin, dysfunction in testosterone levels has long been implicated in aggressive and violent behavior (Archer, 2006). Testosterone is an androgenic steroid found in both men and women, but with much higher concentrations in the former. Among men, however, young adult males tend to show higher rates of aggression and physical violence, such as fighting and homicide, compared to older members of the same sex (Archer, 2004; Daly & Wilson, 1988), especially those that are married with children (Gray, Kahlenberg, Barrett, Lipson, & Ellison, 2002). Differences in the testosterone levels between individuals can be attributed to both social and hereditary factors, but while both animal and human studies suggest that testosterone levels are linked to aggressive acts, such as fighting, assault, and dominance behavior, the role testosterone plays in aggression remains a polarized debate in the social science literature (Archer, 1991). However, despite the controversy over the existence of a direct link between criminality and testosterone, high levels of testosterone have been reported in populations of aggressive individuals, such as criminals with personality disorders, alcoholics, and spousal abusers (Siever, 2008, p. 435). Testosterone appears to play a decisive role in competition as well. A study of competing males found that ‘‘aggressive behavior and change in testosterone concentrations predicted willingness to reengage in another competitive task’’ (Carre´ & McCormick, 2008, p. 408). Salvador and colleagues posit that accompanying this desire for competition associated with increased testosterone levels are psychological variables that increase performance success, such as a high self-confidence and a motivation to win.
Brain Imaging and Political Behavior
239
This in turn, according to the authors, suggests ‘‘an adaptive psychobiological response to competition’’ (Salvador, Suay, Gonzalez-Bono, & Serrano, 2003). Likewise, within the context of a simulated crisis game, a study by McDermott and colleagues found that ‘‘high-testosterone subjects are much more likely to engage in unprovoked attacks against their opponents than their lower-testosterone counterparts’’ (McDermott, Johnson, Cowden, & Rosen, 2007). Furthermore, while the female participants were just as likely as the men to fight back once they were provoked, men were much more likely to initiate a conflict. Through the integration of behavioral endocrinology and cognitive neuroscience, recent brain imaging studies have provided insights into the neural mechanisms associated with the relationship between testosterone and aggression, particularly how the endocrine and neural systems interact to regulate aggressive behavior. Mehta and Beer have shown that testosterone increases aggressive behavior by reducing the ability of the medial OFC to govern self-regulation and impulse control following social provocation (Mehta & Beer, 2010). In addition to affecting the OFC, a study by Birgit Derntl and colleagues found a correlation between testosterone levels and amygdala activation (Derntl et al., 2009). Similar to its influence on the OFC, increased levels of testosterone improved the amygdala’s ability to process threat-related stimuli, especially during episodes of anger and fear. Essentially, the higher the testosterone level, the faster, with more accuracy, the response. These findings in turn suggest that ‘‘when confronted with human facial expressions, testosterone prepares females and males for further behavioral action by enforcing more automatic and autonomic processes leading to attentional shifts and decrease of subconscious fear thereby facilitating approach behavior’’ (Derntl et al., 2009, p. 691).
MAPPING IN-GROUP/OUT-GROUP RELATIONS As we begin to better understand the neural systems and biological mechanisms, such as hormones and neurotransmitters, involved with social behavior, traditional models of political and military decision-making appear limited in their ability to explain intergroup conflict and cooperation. According to a rationalist approach to decision-making, aggressive and violent actions, such as assault and warfare, are the product of costbenefit calculations, that is, if the benefit achieved from attacking outweighs the cost, an aggressive strategy is more likely to be taken. Rational choice
240
JOHN M. FRIEND AND BRADLEY A. THAYER
models have been used by economists to explain the various outcomes of social behavior, from day-to-day negotiations and decision-making to racial discrimination and criminal behavior (Becker, 1993) and scholars of international relations to account for the causes of war and interstate conflict (Waltz, 1979; Mearsheimer, 2003; Jervis, 1978; Van Evera, 1999; Walt, 1987). While the rationalist approach to in-group/out-group relations offers important insights into human political behavior, Fearon correctly points out that the rational actor model fails to adequately address the ‘‘central puzzle,’’ that is, ‘‘war is costly and risky, so rational states should have incentives to locate negotiated settlements that all would prefer to the gamble of war’’ (Fearon, 1995, p. 380). Simply put, it is more rational and cost effective, in blood and treasure, to avoid conflict and make mutually beneficial settlements than to wage war. Unfortunately, this is not the case; history and current events have shown us that war and conflict are the norm of international politics, not the exception. This is not to say that costbenefit calculations are not involved in intergroup conflict, but instead suggests that additional causal mechanisms are responsible for coalitional aggression and cooperation.
PREJUDICE AND DISCRIMINATION For political psychologists and social neuroscientists, a more accurate explanation of the ‘‘central puzzle’’ requires an account of ‘‘people’s cognitive, affective, and behavioral reactions to people from other groups’’ (Fiske, 2002, p. 123). According to Susan Fiske, to understand the automatic behavior that can lead to intergroup conflict, such as prejudice, discrimination, fear, and mistrust, it is necessary to uncover the neural and biological mechanisms that trigger automatic biases. Implicit Association Tests combined with brain imaging studies of the amygdala have shown that intergroup bias occurs automatically under minimal conditions among relatively unprejudiced people (Fiske, 2002; Ashburn-Nardo, Voils, & Monteith, 2001; Ronquillo et al., 2007). Bias in intergroup relations is an automatic response across populations because, as Tooby and Cosmides point out, coalitional aggression and prejudice (against different racial and ethnic groups) are evolutionary traits that improved overall fitness by enabling members of a coalition to gain access to competitive ‘‘reproductive enhancing resources’’ and detect coalitions and alliances (Tooby & Cosmides, 1988; Kurzban, Tooby, & Cosmides, 2001).
Brain Imaging and Political Behavior
241
Before the advent of neural techniques like fMRI and EEG, it was extremely difficult or impossible to identify the mechanisms underlying automatic biases that produce stigmas and stereotypes. Due to advances in brain imaging and neuropsychological methods, we are now able to ‘‘explore the role of specific neural regions and systems in complex social psychological phenomena such as a person’s perceptions and racial prejudice’’ (Derks, Inzlicht, & Kang, 2008, p. 164). Through an examination of the cognitive, affective, and behavioral structures associated with stereotyping and xenophobic responses, we can begin to build a more accurate theory of how these structures influence people’s beliefs and expectations about out-groups (Ochsner & Lieberman, 2001). To better understand the role prejudice and discrimination play in coalitional warfare, such as ethnic conflict, Harris and Fiske point out that it is necessary to acknowledge that not all prejudices are equal (Harris & Fiske, 2006). Unfortunately, most neuroscience studies of automatic bias in intergroup relations contextualize discrimination homogenously and ignore or overlook the effect of emotions on the formation of stereotypes regarding out-groups. Evidence from brain imaging studies of extreme out-group bias reveals the possibility of severe prejudice that can dehumanize targets and possibly lead to genocidal actions. Incorporation of the stereotype content model (SCM) into a neuroscientific approach provides a deeper understanding of the emotions involved with prejudice toward social groups. Harris and Fiske posit that this approach to in-group/out-group relations can offer four distinct emotional responses within the contexts of friend–foe judgment (level of warmth) and capability judgment (level of competence): pride, envy, pity, and disgust (Harris & Fiske, 2006; Fiske, Cuddy, Glick, & Xu, 2002). For example, some groups stereotyped as highly competent and warm, ‘‘elicit the in-group emotions of pride and admiration’’ (Harris & Fiske, 2006, p. 852), while, on the other end of the spectrum, those stereotyped as significantly incompetent with little compassion (or warmth) evoke absolute disgust to the extent that the group members are viewed as less human (or are completely dehumanized). It is with this latter discriminatory categorization and the emotional responses it evokes that the conditions for severe violence and brutality against a specific population or small group of people are made possible, that is to say ‘‘the all-too human ability to commit atrocities such as hate crimes, prisoner abuse, and genocide against people who are dehumanized’’ (Harris & Fiske, 2006, p. 847). At the neuronal level, fMRI scanning shows that extreme discrimination and disgust not only evoke significant response in the amygdala (signifying fear and threat), but also decreased activation of
242
JOHN M. FRIEND AND BRADLEY A. THAYER
the medial prefrontal cortex (mPFC). Since mPFC activation has been observed in ‘‘social cognition tasks in which participants form an impression of a person, rather than an object,’’ little to no activity in the mPFC during interactions with a specific group possibly suggests that the individuals of the target group are valued more as objects than humans (Mitchell, Banaji, & Macrae, 2005). Taking into account the increasing number of ethnic conflicts, some of which have resulted in ethnic cleansing or genocide, in the post-Cold War security environment (Brown, Cote´, Lynn Jones, & Miller, 1997; Thayer, 2009), there indeed appear to be varying degrees of dehumanization during intergroup conflict. Furthermore, as stated earlier, evidence from brain imaging studies suggests that the amygdala also plays a role in perception and behavioral responses to individuals of a different race or ethnicity. In a study by Allen Hart and his colleagues on how perceptions of out-group members differ from in-group members, white and black participants viewed photographs of unfamiliar white and black faces while undergoing fMRI. For all participants, ‘‘the rate of response habituation within the amygdala to face stimuli was dependent upon an interaction between the race of the subjects and the perceived race of the face stimuli’’ (Hart et al., 2000, p. 2353). These results suggest that the amygdala exhibits greater responses to unfamiliar, and possibly threatening, faces, than to familiar faces. Given the amygdala’s role of ‘‘relevance detector’’ (Sander, Grafman, & Zalla, 2003), which includes, but is not limited to, fear-related stimuli, the prejudice and anxiety that occurs between in-group and out-group members during initial interactions can lessen over time through prolonged exposure or, as a study by Kurzban and colleagues found, by manipulating coalition formation and reestablishing coalition membership across racial cleavages (Kurzban et al., 2001). It is important to note, however, that even though evidence suggests that race is an evolutionary byproduct that allowed our ancestors to distinguish alliances and detect threats, discrimination can occur between members of ingroups as well. For example, an fMRI study of the amygdala’s sensitivity to race revealed that both African-American and Caucasian-Americans showed greater amygdala response to black faces than to white faces (Lieberman, Hariri, Jarcho, Eisenberger, & Bookheimer, 2005). Since amygdala activity in African-American participants exhibited greater response to the black target faces, the authors speculate that ‘‘the amygdala activity typically associated with race-related processing may be a reflection of culturally learned negative associations regarding African-American individuals’’ (Lieberman et al., 2005, p. 722). In other words, intergroup prejudice and discrimination is clearly a biological and cultural phenomenon.
Brain Imaging and Political Behavior
243
Along these lines, a recent study on ‘‘social cognition in people from different cultural backgrounds’’ found increased activation in the precuneus (PC) among Israeli and Arab participants as they read pro-out-group versus pro-in-group statements (Bruneau & Saxe, 2010). Activation in the PC is believed to signify emotional reasoning during difficult moral judgments of harmful behavior (Greene, Sommerville, Nystrom, Darley, & Cohen, 2001), which suggests, according to the authors, ‘‘strong ingroup bias in evaluating the reasonableness of partisan statements about the Middle East’’ (Bruneau & Saxe, 2010, p. 1709). Although these findings support previous research on neuronal and evolutionary reasons for prejudice and stereotypes in intergroup relations, the particularly high level of in-group bias among Israelis and Arabs is undoubtedly exacerbated by the politically hostile context that those participants interact within on a daily basis, which in turn can make extremely aggressive conflict between such groups significantly more likely and harder to mitigate.
ALTRUISM AND COOPERATION As Tooby and Cosmides point out, both coalitional aggression and alliance formation not only suggest the development of neural circuits that allow ingroups to detect threats, but also the need for cognitive mechanisms that enable the coalition to ‘‘coalesce, function, and sustain itself as a group of cooperating individuals’’ (Tooby & Cosmides, 1988, p. 2). Through the use of brain imaging techniques, the fields of cognitive neuroscience and behavioral genetics, among others, have shed light on the neuronal substrates and biological mechanisms involved with altruism and cooperation within and between groups. For many scholars of international politics, improving transparency and dialogue among state and nonstate actors has been seen as a viable strategy for promoting cooperation and mitigating conflict within an anarchic international system. Most notably, international institutionalists, such as Robert Keohane and Stephen Krasner, assert that international organizations and regimes, such as the United Nations and the General Agreement on Tariffs and Trade (GATT), can create clearly established structures, principles, rules, norms, and decision-making procedures that foster interdependence and create a positive sum context for cooperative decision-making to take place (Keohane, 1998; Krasner, 1983). While international institutions and regimes have had their fair share of successes, there have also been countless failures to promote cooperation and stability
244
JOHN M. FRIEND AND BRADLEY A. THAYER
that resulted in conflict and war (Mearsheimer, 1994/1995). This difficulty with fostering collaboration and cooperation suggests that establishing trust and building altruistic relationships require more than simply manipulating rational cost-benefit calculations. Likewise, if human decision-making during intergroup relations has evolved to be zero sum, how reliable are the rationalist assumptions of international diplomacy for mitigating xenophobia, mistrust, and prejudice? An evolutionary approach to in-group altruism proposes that lethal intergroup competition and selective group murder allowed for altruistic behaviors to evolve in in-group relations since groups with more altruistic members survive during coalitional warfare (Bowles, 2006). Along similar lines, Choi and Bowles posit that altruism became an adaptive trait because an individual adopting the behaviors of intergroup hostility and aggression incur ‘‘mortal risks or foregoes beneficial opportunities for coalitions, coinsurance, and exchange, thereby incurring a fitness loss by comparison to those who eschew hostility towards other groups’’ (Choi & Bowles, 2007, p. 636). For the authors, this form of altruistic behavior can be seen as an example of parochial altruism, which they argue survived natural selection because, within a resource competitive environment, coalitions with many parochial altruists, that is, those willing to aggress against member of outgroups, were more likely to survive conflicts. Furthermore, as McDermott points out, since ‘‘political coalitions rest on a foundation of distinguishing in-group and out-group members’’ (McDermott, 2009, p. 573), coalition formation and cooperation require a process of social inclusion and exclusion, that is to say, an identification of us versus them, for overall survival and fitness maximization. To explain the frequency of in-group altruism, especially in instances of cooperation with genetically unrelated strangers with little or no reputation gain, Fehr and Rockenbach posit that an evolutionary model suggests that ‘‘strong reciprocity, the behavioral propensity for altruistic punishment and altruistic reward, is of key importance for human cooperation’’ (Fehr & Rockenbach, 2004, p. 784). Moreover, in contrast to the weak reciprocator associated with reciprocal altruism, a strong reciprocator is argued to ‘‘cooperate with others and punish non-cooperators, even when this behavior cannot be justified in terms of self-interest, extended kinship, or reciprocal altruism’’ (Gintis, 2000, p. 169). While this understanding of altruistic behavior falls outside the explanatory power of the rational actor model, Gintis argues that strong reciprocity can be seen in various social and political actions, such as the time and energy spent by victims of crime to ensure that the perpetrators are apprehended and punished as well as the all too common
Brain Imaging and Political Behavior
245
desire of humans to ‘‘engage in episodic collective action towards transforming social norms and political regimes’’ (Gintis, 2000, pp. 177–178). While Darwin was one of the first to recognize that ‘‘groups with a greater number of courageous, sympathetic and faithful members, who were always ready to warn each other of danger, to aid and defend each other y would spread and be victorious over other tribes’’ (Darwin, 1873, p. 156), recent economic studies using fMRI have provided experimental evidence that supports the strong reciprocity hypothesis. During neuroeconomic experiments of decision-making during an ultimatum game (Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003), activation of the bilateral insula, the area of the brain associated with pain and disgust, the dorsolateral prefrontal cortex (DLPFC), a region linked to goal maintenance, and the ACC, the area implicated in identification of cognitive conflict, occurred when participants received unfair offers (Fehr & Rockenbach, 2004, pp. 786–787). Fehr and Rockenbach conclude that the involvement of the anterior insula suggests a ‘‘willingness to reject unfair outcomes’’ while activation of the reward-related circuits of the DLPFC and ACC ‘‘seem to be important for human cooperation and the punishment of norm violations’’ (Fehr & Rockenbach, 2004, p. 789). Building from the findings of brain imaging studies on parochial altruism, Carsten De Dreu and colleagues posit that the neurotransmitter and hormone oxytocin modulates parochial altruistic behaviors in situations when humans are arranged into in-groups and competing out-groups (De Dreu et al., 2010). During intergroup prisoner’s dilemma-maximizing difference (IPD-MD) games, participants given oxytocin, in contrast to those given a placebo, displayed more accounts of in-group trust and in-group love, as well as the tendency to exhibit defensive aggression toward the competing out-group. In a similar study, Michael Kosfeld and his colleagues found that participants administered doses of oxytocin during trust games with real monetary stakes not only displayed higher levels of trust, but also a significant ‘‘willingness to accept social risks arising through interpersonal interactions’’ (Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, 2005 p. 673).
MAPPING PARTISAN PREFERENCE AND POLITICAL JUDGMENT Brain imaging studies of the neural bases of partisan political judgment, preference, and voting can be seen as some of the most insightful and
246
JOHN M. FRIEND AND BRADLEY A. THAYER
controversial research on political behavior. The factors influencing voting behavior and turnout have been a long time interest of political scientists. Much of the recent research using brain imaging techniques to explain voting and other forms of political participation strongly suggest that neural substrates and biological mechanisms shape our political orientations, preferences, and decisions regarding candidates. For example, many researchers have recognized that the visual appearance of the candidate is an important factor shaping voter judgments. A classic example is the 1960 first presidential debate between John F. Kennedy and Richard Nixon. Polls following the debate revealed that television viewers considered Kennedy to be the winner while radio listeners, on the other hand, thought Nixon won the debate. To account for this discrepancy among viewers and listeners, many have suggested that the appearance of the candidates during the debate influenced the opinions of the television viewers. On this point, Stanton writes: Kennedy was bronzed beautifully, wearing a navy suit and a blue shirt. Nixon looked like death because he had been in the hospital. And you could run your hand inside his collar without touching anything – it was that loose. His color was terrible; his beard was not good and he didn’t want any makeup. (Stanton, 2000)
While previous research on voting behavior has identified that a candidate’s appearance, such as attractiveness, body type, and temperament, shape voter opinion, recent cognitive neuroscience studies using brain imaging techniques provide us with a better understanding of how emotion and motivated reasoning shape political judgment and partisan preference. Much of the research on how appearance affects voting behavior focuses on the facial expressions of candidates and what regions of the brain respond to the emotional displays. Traditional approaches to the processes in which voters identify with political candidates tend to ignore the affect effect and instead conclude that voters are more responsive to the candidates’ position on the issues, their party affiliation, and physical traits like race, gender, and age. Although these factors undoubtedly influence, to varying degrees, voter opinion, to leave out the influence of affect from the study of voter preference is to ignore the central role emotion plays in mediating attitude change. Denis Sullivan and Roger Masters point out that ‘‘emotional reactions to facial displays and other nonverbal behaviors may play an important role in producing ‘momentum’ some candidates develop’’ in the sense that as voters respond ‘‘to these displays with emotions of happiness and warmth, subsequent attitudes toward the candidate are likely to be more positive’’ (Sullivan & Masters, 1988, p. 361).
Brain Imaging and Political Behavior
247
For example, Masters argues that conventional theories of public opinion are unable to account for the preferences of Republican voters during the 1992 presidential campaign (Masters, 2001b). The emotional responses of voters to the facial displays of George Bush, Dan Quayle, and Patrick Buchannan unexpectedly did not follow party lines. For Republican women voters, the happy/reassurance displays of the candidates evoked more feelings of warmth than Democratic women, but Republican men, on the other hand, reported less warmth than the men affiliated with the Democratic Party. Masters suggests that these findings challenge the assumptions of conventional stimulus–response models, which speculate that the effects of factors such as gender and party identification are additive, by revealing ‘‘some voters may integrate information in entirely different ways, so that frequency of watching television or gender can modify the intensity and even the valence of emotional reactions’’ (Masters, 2001b, p. 70). Brain imaging techniques have intersected with behavioral studies on the psychological mechanisms involved with appearance-based judgments to provide valuable information on the underlying neural processes of voter preference. In one such study, Michael Spezio and colleagues conducted two separate fMRI experiments, a simulated voting study and a social judgment study, in which two separate groups of participants viewed pairs of unfamiliar real politicians (both Republican and Democrat) from previous elections (Spezio et al., 2008). In the first study, participants were asked to vote on the candidates following a brief display of their facial images, while, in the second, the participants were requested to comment on the appearances (threatening, attractive, deceitful, or competent) of a completely different pair of politicians. In both cases, each group of participants exhibited activity in the insula and anterior cingulate, regions associated with processing negatively valenced emotions, during images of candidates who had lost the mock elections. These findings, according to the authors, suggest voters’ decisions to support a candidate are not solely driven by positive motives but also by negative affect, that is to say, ‘‘political ‘intangibles,’ such as a candidate’s appearance, might also work primarily via negative motives’’ (Spezio et al., 2008, p. 349). While numerous studies have shown that a candidate’s appearance can induce both positive and negative affect, it remains less clear what attributes of a person’s appearance signal negative traits that threaten voters and result in election loss. A cross-cultural fMRI study of the influence of the amygdala on political judgments and subsequent voting decisions may hold
248
JOHN M. FRIEND AND BRADLEY A. THAYER
some insight into the involvement of ‘‘political intangibles’’ (Rule et al., 2009). Among 28 American and Japanese university students, participants viewed and made judgments on 58 facial images of candidates from the 2004 and 2006 U.S. Senate elections and 58 candidates from the 2000 election of the Japanese Diet. The study found that with both American and Japanese participants, ‘‘the bilateral amygdala was significantly more responsive to the faces of politicians for whom participants chose to vote’’ and ‘‘participants showed a greater overall amygdala response to the faces of outgroup candidates than they did the faces of ingroup candidates’’ (Rule et al., 2009, pp. 353–354). Since these findings are consistent with studies of the neural underpinnings of prejudice and discrimination involved in evaluating out-group members, it appears that voter judgment and decisionmaking are not only influenced by rational cost-benefit evaluations of a candidate’s issues and policies, but are also, and possibly more so, the product of emotional reactions to adaptive biological mechanisms associated with maintaining in-group cohesion and survival. Along similar lines, Drew Westen and colleagues argue that a neuronal approach to emotion-biased motivated reasoning provides a more accurate understanding of the formation of partisan political judgment (Westen, Blagov, Harenski, Kilts, & Hamann, 2006). Through an fMRI study involving 30 committed partisans (15 Democrats and 15 Republicans) during the 2004 presidential election, participants were presented with threatening information regarding their party’s candidate, the opposing party’s candidate, and neutral control targets. When evaluating emotionally threatening information about a preferred candidate (Republicans presented with George Bush contradictions and Democrats presented with John Kerry contradictions), participants displayed activations of the ventromedial prefrontal cortex, ACC, posterior cingulate cortex, insular cortex, and lateral orbital cortex. Since these areas of the brain are associated with punishment, pain, negative affect, distress-related to error detection, and experiences of judging forgivability of actions and emotionally laden moral judgments, the authors suggest that decisions regarding political candidates are made under severe emotional constraints that produce emotional-biased choices. Interestingly, during the experiments, the participants exhibited little to no differential activation of the DLPFC, the region of the brain associated with ‘‘cold reasoning’’ and ‘‘explicit emotion suppression,’’ which in turn suggests the primacy of emotion, rather than rational thought, during the formation of partisan political judgments.
Brain Imaging and Political Behavior
249
THE LIMITATIONS OF BRAIN IMAGING TECHNIQUES Contemporary brain imaging studies and techniques have offered a wealth of new information about human political behavior that has already begun to challenge conventional models of conflict, coalition formation, decisionmaking, and public opinion and behavior. While the fields of cognitive neuroscience and behavioral genetics will undoubtedly continue to shape social science research and methodology in the years to come, it is important to note that the application of brain imaging techniques to the study of social and political phenomena have their limitations and for good reason. The first major concern is centered on how to interpret the rate of brain activity, that is, the blood oxygen level dependent (or BOLD) signal. Rose McDermott point outs that ‘‘any given moment of activity does not offer much information about the circuits of activity that preceded or followed the image that is captured in time on the brain scan’’ (McDermott, 2009, pp. 578–579). By taking this limitation into account, McDermott is recommending a parsimonious interpretation of the cognitive meanings of the BOLD signal, that is to say, brain imaging techniques should not be used solely to create an activation map of functionality, but instead serve as a geographical approach that incorporates psychological and neurobiological theories to explain the relationship between cognitive and behavioral responses (McDermott, 2009). Second, although it is a common practice in fMRI studies to standardize the brains of participants, a method referred to as ‘‘spatial normalization,’’ it is important, however, to recognize that ‘‘individual brains typically are not identical in shape, especially cortical regions, or, more important, in patterns of localized activation’’ (Spezio & Adolphs, 2007). Michael Spezio and Ralph Adolphs argue that such normalization can overlook ‘‘potentially important empirically determined associations between information processing and brain activity’’ (Spezio & Adolphs, 2007) and therefore must be taken into consideration. Finally, brain imaging should be seen as a methodological approach, or tool, that promotes interdisciplinary collaboration between neuroscientists and social scientists and not a technique that holds all the answers to questions regarding human social and political behavior. It would be grossly misguided to ignore how culture and environmental stimuli shape human interactions or assume that human nature is purely a product of the neuronal content of our brains. ‘‘Most aspects of the mind y require a more
250
JOHN M. FRIEND AND BRADLEY A. THAYER
comprehensive approach to reveal the mystery of mind-brain connections,’’ and only through a combination of the ‘‘emerging knowledge and new techniques from the neurosciences and cognitive sciences with expertise in social psychology’’ will a truly accurate understanding of human political behavior be achieved (Cacioppo et al., 2003).
CONCLUSION: THE FUTURE OF BRAIN IMAGING Our chapter has identified how the science of the brain will make major contributions to the discipline of political science. Specifically, we have argued brain imagining will greatly assist studies of aggression and violence; analysis of prejudice and in-group/out-group formation; and the examination of voting behavior and partisan preference. Despite these significant contributions, we recognize that these contributions are only the start. In fact, the brain sciences have the ability to remake the discipline of political science. We believe that, indeed, this will happen as neuroscience continues to advance. Reflecting on the discipline broadly, political science has traditionally focused on identifying causation anywhere but the individual: organizational approaches, the study of bureaucracy, systemic analyses, all have been favored since the triumph of behavioralism in the 1950s. Almost forced out of the discipline were the studies of individuals and psychological approaches. With the advance of science, we will see the renaissance of individual approaches: the first image, or individual level of analysis for causation of political behavior will be revived. Over time, the consequences of this will transform the balance of power in the discipline and in departments. Rational choice will wane, political psychology will wax as the center of the discipline. The glorification of economics and economic approaches will fade too. It is ironic that psychology was once dismissed as the ‘‘softest’’ of the social sciences, while economics was triumphed as the most scientific. Yet the advance of the life sciences, evolutionary psychology, cognitive psychology and neuroscience altered this ordering. With the decline of economics as a model, the tenuous relationship political science has had with psychology will grow stronger as political scientists adopt their technology and approaches to the study of individual political behavior. The result for the creation of political knowledge will be positive, and new avenues of research will be opened for political scientists – a positive development for any academic discipline. Political scientists may
Brain Imaging and Political Behavior
251
come to appreciate Aristotle’s observation that man is political animal in a new light. Understandably, the political has been touted thus far. In the future, the animal will be as the study of the biological revolutionizes our understanding of the political.
REFERENCES Alford, J. R., & Hibbing, J. R. (2004). The origin of politics: An evolutionary theory of political behavior. Perspectives on Politics, 2(4), 707–723. Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., & Damasio, A. (1999). Impairment of social and moral behavior related to early damage in human prefrontal cortex. Nature Neuroscience, 2(11), 1032–1037. Anon. (2007). Mind games: How not to mix politics and science. Nature, 450(7169), 457. Archer, J. (1991). The influence of testosterone on human aggression. British Journal of Psychology, 82(1), 1–28. Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-analysis. Review of General Psychology, 8(4), 291–322. Archer, J. (2006). Testosterone and human aggression: An evaluation of the challenge hypothesis. Neuroscience and Biobehavioral Reviews, 30(3), 319–345. Arnsten, A. (1998). The biology of being frazzled. Science, 280(5370), 1711–1712. Ashburn-Nardo, L., Voils, C. I., & Monteith, M. J. (2001). Implicit associations as the seeds of intergroup bias: How easily do they take root? Journal of Personality and Social Psychology, 81(5), 789–799. Barber, C. (2008). The brain: A mindless obsession. The Wilson Quarterly, 32(1), 32–44. Becker, G. (1993). Nobel lecture: The economic way at looking at behavior. The Journal of Political Economy, 101(3), 385–409. Blank, R. H. (1999). Brain policy: How the new neuroscience will change our lives and our politics. Washington, DC: Georgetown University Press. Blank, R. H. (2006). The brain, aggression, and public policy. Politics and the Life Sciences, 24(1-2), 12–21. Bowles, S. (2006). Group competition, reproductive leveling, and the evolution of human altruisms. Science, 314(5805), 1569–1572. Brown, M., Cote´, O., Jr., Lynn Jones, S., & Miller, S. (Eds). (1997). Nationalism and ethnic conflict: An international security reader. Cambridge, MA: MIT Press. Bruneau, E. G., & Saxe, R. (2010). Attitudes towards the outgroup are predicted by activity in the precuneus in Arabs and Israelis. NeuroImage, 52(4), 1704–1711. Cacioppo, J. T., Berntson, G. G., Lorig, T. S., Norris, C. J., Rickett, E., & Nusbaum, H. (2003). Just because you’re imaging the brain doesn’t mean you can stop using your head: A primer and set of first principles. Journal of Personality and Social Psychology, 85(4), 650–661. Cacioppo, J. T., Berntson, G. G., & Nusbaum, H. C. (2008). Neuroimaging as a new tool in the toolbox of psychological science. Current Directions in Psychological Science, 17(2), 62–67. Carre´, J. M., & McCormick, C. M. (2008). Aggressive behavior and change in salivary testosterone concentrations predict willingness to engage in a competitive task. Hormones and Behavior, 54(3), 403–409.
252
JOHN M. FRIEND AND BRADLEY A. THAYER
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., Taylor, A., & Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297(5582), 851–854. Changeux, J. P. (1997). Neuronal man: The biology of mind. Princeton, NJ: Princeton University Press. Choi, J. K., & Bowles, S. (2007). The coevolution of parochial altruism and war. Science, 318(5850), 636–640. Confer, J. C., Easton, J. A., Fleischman, D. S., Goetz, C. D., Lewis, D. M. G., Perilloux, C., & Buss, D. (2010). Evolutionary psychology: Controversies, questions, prospects, and limitations. American Psychologist, 65(2), 110–126. Crick, F. (1979). Thinking about the brain. Scientific American, 241, 219–232. Daly, M., & Wilson, M. (1988). Homicide. New York: Aldine de Gruyter. Darwin, C. (1873). The descent of man. New York: Appleton. Davidson, R. J., Putnam, K. M., & Larson, C. L. (2000). Dysfunction in the neural circuitry of emotion regulation: A possible prelude to violence. Science, 289(5479), 591–594. De Dreu, C. K. W., Greer, L. L., Handgraaf, M. J. J., Shalvi, S., Van Kleef, G. A., Baas, M., Ten Velden, F. S., Van Dijk, E., & Feith, S. W. W. (2010). The neuropeptide oxytocin regulates parochial altruism in intergroup conflict among humans. Science, 328(5984), 1408–1411. Derks, B., Inzlicht, M., & Kang, S. (2008). The neuroscience of stigma and stereotype threat. Group Processes & Intergroup Relations, 11(2), 161–181. Derntl, B., Windischerger, C., Robinson, S., Kryspin-Exner, I., Gur, R. C., Moser, E., & Habel, U. (2009). Amygdala activity to fear and anger in healthy young males is associated with testosterone. Psychoneuroendocrinology, 34(5), 687–693. Fearon, J. D. (1995). Rationalist explanations for war. International Organizations, 49(3), 379–414. Fehr, E., & Rockenbach, B. (2004). Human altruism: Economic, neural, and evolutionary perspectives. Current Opinion in Neurobiology, 14(6), 784–790. Fiske, S. T. (2002). What we know about bias and intergroup conflict, the problem of the century. Current Directions in Psychological Science, 11(4), 123–128. Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878–902. Freud, S. (1922). Beyond the pleasure principle. C. J. M. Hubback (Trans.). London, UK: The International Psycho-Analytical Press. Gintis, H. (2000). Strong reciprocity and human sociality. Journal of Theoretical Biology, 206(2), 169–179. Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. L., & Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23(3), 193–201. Greene, J. D., Sommerville, B. R., Nystrom, L. E., Darley, J. M., & Cohen, J. D. (2001). An fMRI Investigation of emotional engagement in moral judgment. Science, 293(5537), 2105–2108. Harris, L. T., & Fiske, S. T. (2006). Dehumanizing the lowest of the low: Neuroimaging responses to extreme out-groups. Psychological Science, 17(10), 847–853. Hart, A. J., Whalen, P. J., Shin, L. M., McInerney, S. C., Fischer, H., & Rauch, S. L. (2000). Differential response in the human amygdala to racial outgroup vs ingroup face stimuli. NeuroReport, 11(11), 2351–2354.
Brain Imaging and Political Behavior
253
Iacoboni, M., Freedman, J., Kaplan, J., Jamieson, K. H., Freedman, T., Knapp, B., & Fitzgerald, K. (2007). This is your brain on politics. New York Times. Available at http:// www.nytimes.com/2007/11/11/opinion/11freedman.html?_r ¼ 1 Jervis, R. (1978). Cooperation under the security dilemma. World Politics, 30(2), 167–214. Kalin, N. H. (1999). Primate models to understand human aggression. Journal of Clinical Psychiatry, 60(Suppl. 15), 29–32. Keeley, L. (1997). War before civilization: The myth of the peaceful savage. Oxford, UK: Oxford University Press. Keohane, R. O. (1998). International institutions: Can interdependence work? Foreign Policy, 110, 82–96. Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435(7042), 673–676. Kotulak, R. (1997). Inside the brain: Revolutionary discoveries of how the mind works. Kansas City, MO: Andrews McMeel. Krasner, S. D. (Ed.) (1983). International regimes. Ithaca, NY: Cornell University Press. Kruesi, M. J. P., Hibbs, E. D., Zahn, T. P., Keysor, C. S., Hamburger, S. D., Bartko, J. J., & Rapoport, J. L. (1992). A 2-year prospective follow-up study of children and adolescents with disruptive behavior disorders: Prediction by cerebrospinal fluid 5-hydroxyindoleacetic acid, homovanillic acid, and autonomic measures?. Archives of General Psychiatry, 49(6), 429–435. Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Science, 98(26), 15387–15392. LeDoux, J. (1995). Emotion: Clues from the brain. Annual Review of Psychology, 46, 209–235. Lieberman, M. D., Hariri, A., Jarcho, J. M., Eisenberger, N. I., & Bookheimer, S. Y. (2005). An fMRI investigation of race-related amygdala activity in African-American and Caucasian-American individuals. Nature Neuroscience, 8(6), 720–722. London, E. D., Ernst, M., Grant, S., Bonson, K., & Weinstein, A. (2000). Orbitofrontal cortex and human drug abuse: Functional imaging. Cerebral Cortex, 10(3), 334–342. Lorenz, K. (1974). On aggression. M. K. Wilson (Trans.). New York: Mariner Books. Masters, R. D. (2001a). Biology and politics: Linking nature and nurture. Annual Review of Political Science, 4, 345–369. Masters, R. D. (2001b). Cognitive neuroscience, emotion, and leadership. In: J. H. Kuklinkski (Ed.), Citizens and politics: Perspectives from political psychology (pp. 68–102). Cambridge, UK: Cambridge University Press. Masters, R. D., & McGuire, M. T. (Eds). (1994). The neurotransmitter revolution: Serotonin, social behavior, and the law. Carbondale, IL: Southern Illinois University Press. McDermott, R. (2009). Mutual interests: The case of increasing dialogue between political science and neuroscience. Political Research Quarterly, 62(3), 571–583. McDermott, R., Johnson, D., Cowden, J., & Rosen, S. (2007). Testosterone and aggression in a simulated crisis game. The Annals of the American Academy of Political and Social Science, 614, 15–33. Mearsheimer, J. (1994/1995). The false promise of international institutions. International Security, 19(3), 5–49. Mearsheimer, J. (2003). The tragedy of great power politics. New York: W.W. Norton & Company.
254
JOHN M. FRIEND AND BRADLEY A. THAYER
Mehta, P. H., & Beer, J. (2010). Neural mechanisms of the testosterone-aggression relation: The role of orbitofrontal cortex. Journal of Cognitive Neuroscience, 22(10), 2357–2368. Miller, G. (2008). Neuroimaging: Growing pains for fMRI. Science, 320(5882), 1412–1414. Mitchell, J. P., Banaji, M. R., & Macrae, C. N. (2005). The link between social cognition and self-referential thought in the medial prefrontal cortex. Journal of Cognitive Neuroscience, 17(8), 1306–1315. Mozes, A. (2001). Violent images activate brain region: Study. Reuters. Ochsner, K. N., & Lieberman, M. D. (2001). The emergence of social cognitive neuroscience. American Psychologist, 56(9), 717–734. Paus, T. (2005). Mapping brain development and aggression. The Canadian Child and Adolescent Psychiatry Review, 14(1), 10–15. Raichle, M. E. (2003). Social neuroscience: A role for brain imaging. Political Psychology, 24(4), 759–764. Ronquillo, J., Denson, T. F., Lickel, B., Lu, Z. L., Nandy, A., & Maddox, K. B. (2007). The effects of skin tone on race-related amygdala activity: An fMRI investigation. SCAN, 2(1), 39–44. Rosen, S. P. (2005). War and human nature. Princeton, NJ: Princeton University Press. Rule, N. O., Freeman, J. B., Moran, J. M., Gabrieli, J. D. E., Adams, R. B., Jr., & Ambady, N. (2009). Voting behavior is reflected in amygdala response across culture. Social Cognitive and Affective Neuroscience, 5(2-3), 349–355. Salvador, A., Suay, S. F., Gonzalez-Bono, E., & Serrano, M. A. (2003). Anticipatory cortisol, testosterone and psychological responses to judo competition in young men. Psychoneuroendocrinology, 28(3), 364–375. Sander, D., Grafman, J., & Zalla, T. (2003). The human amygdala: An evolved system for relevance detection. Reviews in Neuroscience, 14(4), 303–316. Sanfey, A. G., Rilling, J. K., Aronson, J. A., Nystrom, L. E., & Cohen, J. D. (2003). The neural basis of economic decision-making in the ultimatum game. Science, 300(5626), 1755–1758. Sell, A., Tooby, J., & Cosmides, L. (2009). Formidability and the logic of human anger. Proceedings of the National Academy of Science, 106(35), 15073–15078. Siever, L. J. (2008). Neurobiology of aggression and violence. American Journal of Psychiatry, 165(4), 429–442. Siever, L., & Trestman, R. L. (1993). The serotonin system and aggressive personality disorder. International Clinical Psychopharmacology, 8(Suppl. 2), 33–40. Somit, A. (1968). Toward a more biologically-oriented political science: Ethology and psychopharmacology. Midwest Journal of Political Science, 12(4), 550–567. Spezio, M. L., & Adolphs, R. (2007). Emotional processing and political judgment: Toward integrating political psychology and decision neuroscience. In: W. R. Neuman, G. E. Marcus, A. N. Crigler & M. MacKuen (Eds), The affect effect: Dynamics of emotion in political thinking and behavior (pp. 71–95). Chicago, IL: The University of Chicago Press. Spezio, M. L., Rangel, A., Alvarez, R. M., O’Doherty, J. P., Mattes, K., Todorov, A., Kim, H., & Adolphs, R. (2008). A neural basis for the effect of candidate appearance on election outcomes. SCAN, 3(4), 344–352. Stanton, F. (2000). The first debate over presidential debates. Newsweek. Sullivan, D. G., & Masters, R. D. (1988). Happy warriors: Leaders’ facial displays, viewers’ emotions, and political support. American Journal of Political Science, 32(2), 345–368. Thayer, B. A. (2000). Bringing in Darwin: Evolutionary theory, realism, and international politics. International Security, 25(2), 124–151.
Brain Imaging and Political Behavior
255
Thayer, B. A. (2003). Ethnic conflict and state building. In: A. Somit & A. Peterson (Eds), Human nature and public policy: An evolutionary approach (pp. 225–242). New York: Palgrave Macmillan. Thayer, B. A. (2009). Darwin and international relation: On the evolutionary origins of war and ethnic conflict. Lexington, KY: University Press of Kentucky. Tiihonen, J., Virkkunen, M., R€as€anen, P., Pennanen, S., Sainio, E. L., Callaway, J., Halonen, P., & Liesivuori, J. (2001). Free L-tryptophan plasma levels in antisocial violent offenders. Psychopharmacology, 157(4), 395–400. Tingley, D. (2006). Neurological imaging as evidence in political science: A review, critique, and guiding assessment. Social Science Information, 45(1), 5–33. Tooby, J., & Cosmides, L. (1988). The evolution of war and its cognitive foundations. Institute for Evolutionary Studies Technical Reports 88-1. Tooby, J., & Cosmides, L. (2010). Groups in mind: The coalitional roots of war and morality. In: H. Høgh-Olesen (Ed.), Human morality and sociality: Evolutionary and comparative perspective (pp. 191–234). New York: Palgrave Macmillan. Van Evera, S. (1999). Causes of war: Power and the roots of conflict. Ithaca, NY: Cornell University Press. Virkkunen, M., De Jong, J., Bartko, J., & Linnoila, M. (1993). Psychobiological concomitant of history of suicide attempts among violent offenders and impulsive fire setters. Archives of General Psychiatry, 46(7), 604–606. Walt, S. (1987). The origins of alliances. Ithaca, NY: Cornell University Press. Waltz, K. (1979). Theory of international politics. Reading, MA: Addison-Wesley. Westen, D., Blagov, P. S., Harenski, K., Kilts, C., & Hamann, S. (2006). Neural bases of motivated reasoning: An fMRI study of emotional constraints on partisan political judgment in the 2004 U.S. presidential election. Journal of Cognitive Neuroscience, 18(11), 1947–1958.