Homicide: A Psychiatric Perspective 2nd Edition

HOMICIDE A Psychiatric Perspective Second Edition This page intentionally left blank HOMICIDE A Psychiatric Perspec...

Author: Carl P. | M.D. Malmquist

53 downloads 6902 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

HOMICIDE A Psychiatric Perspective Second Edition

This page intentionally left blank

HOMICIDE A Psychiatric Perspective Second Edition

Carl P. Malmquist, M.D., M.S. Professor of Social Psychiatry, Department of Sociology University of Minnesota, Minneapolis, Minnesota

Washington, DC London, England

Note: The authors have worked to ensure that all information in this book is accurate at the time of publication and consistent with general psychiatric and medical standards, and that information concerning drug dosages, schedules, and routes of administration is accurate at the time of publication and consistent with standards set by the U.S. Food and Drug Administration and the general medical community. As medical research and practice continue to advance, however, therapeutic standards may change. Moreover, specific situations may require a specific therapeutic response not included in this book. For these reasons and because human and mechanical errors sometimes occur, we recommend that readers follow the advice of physicians directly involved in their care or the care of a member of their family. Books published by American Psychiatric Publishing, Inc., represent the views and opinions of the individual authors and do not necessarily represent the policies and opinions of APPI or the American Psychiatric Association. Copyright © 2006 American Psychiatric Publishing, Inc. ALL RIGHTS RESERVED Manufactured in the United States of America on acid-free paper 10 09 08 07 06 5 4 3 2 1 Second Edition Typeset in Adobe’s Futura and Palatino. American Psychiatric Publishing, Inc. 1000 Wilson Boulevard Arlington, VA 22209–3901 www.appi.org Library of Congress Cataloging-in-Publication Data Homicide : a psychiatric perspective / edited by Carl P. Malmquist.— 2nd ed. p. ; cm. Rev. ed. of: Homicide / Carl P. Malmquist. 1st ed. c1996. Includes bibliographical references and index. ISBN 1-58562-204-4 (pbk. : alk. paper) 1. Homicide—Psychological aspects. 2. Murderers—Mental health. [DNLM: 1. Homicide—psychology. 2. Mental Disorders— psychology. WM 140 H767 2006] I. Malmquist, Carl P. II. Malmquist, Carl P. Homicide. RC569.5.H65M35 2006 616.85'844—dc22 2005032512 British Library Cataloguing in Publication Data A CIP record is available from the British Library.

CONTENTS Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Preface to the Second Edition. . . . . . . . . . . . . . . . ix Preface to the First Edition . . . . . . . . . . . . . . . . . . xv Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . .xvii

1

Epidemiological Aspects of Homicide . . . . . . . . . 1

2

Biological Factors in Homicide . . . . . . . . . . . . . . 55

3 4 5 6 7

Schizophrenia, Delusional Disorders, and the Prediction Problem Regarding Homicide . . . . . . . . . . . . . . . . . . . . . . . 91 Borderline Personality Disorder and Homicide: The Quest for Vindication . . . . 121 Dependent Personality Disorder and Homicide . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Narcissism and Homicide. . . . . . . . . . . . . . . . . . 177 Masochism and Homicide: The Ultimate Enslavement . . . . . . . . . . . . . . . . . 199

8

The Depressed or Bipolar Person and Homicide . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

9

Juveniles and Homicide . . . . . . . . . . . . . . . . . . . 283

10

Sexual Homicide . . . . . . . . . . . . . . . . . . . . . . . . . 331

11 12

Legal Versus Clinical Views on Homicide: Diagnosis and Voluntariness. . . 363 Homicide in the Twenty-First Century: Where Knowledge Is Needed . . . . . . . . . . . . . . 393 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419

FOREWORD The heart has its reasons which reason does not understand.

—Blaise Pascal Newspapers and television hawk a steady stream of homicides. Local television news brings neighborhood homicides into viewers’ homes. Movies, books, plays, and documentaries are replete with homicide depictions. The average American by age 18 years will have viewed 250,000 acts of violence, including 40,000 murders on television. The public has an unquenchable thirst for murder mysteries, real or fictional. Murders sell papers and advertised products. Homicide is a multimillion-dollar media industry. But why is there such an enormous interest in an unfortunately common tragedy? As a forensic psychiatrist, I am convinced that the public’s interest in homicide derives from a deep, dark sense that the capacity for murder is quintessentially human. Blood has stained the fabric of human history. Wars, genocides, inquisitions, and pogroms are well known, despite their deniers. Since Adam and Eve’s fall from grace, the history of the world has been filled with violence. Today is no different. It is evident that many people are both fascinated and appalled by humanity’s dark side. Serial killers especially invoke these feelings by their spectacularly horrendous crimes. People protest and reject the notion that the difference between individuals who are labeled as “bad” or praised as “good” is not one of kind but one of degree. Theodore Reich, a pioneering psychoanalyst, observed, “If wishes were horses, they would pull the hearses of our dearest friends and nearest relatives. All men are murderers at heart.” This sentiment, of course, is not news. The biblical prophet Jeremiah lamented, “The heart is deceitful above all things, and desperately wicked; who can know it?” But study it we can. There is no reason for despair. In this book, Dr. Carl Malmquist, a preeminent psychiatrist and forensic psychiatrist, examines homicide through the lens of his experience, his knowledge, and the latest research. The chapters on epidemiology and biological substrata of homicide are disturbingly informative. The role of mental illness

vii

viii

HOMICIDE: A PSYCHIATRIC PERSPECTIVE

in homicide always seems to raise the question of whether the murderer was “mad” or “bad” in insanity defense cases. Here, forensic psychiatry and the law intersect. The application of psychiatry to the law, however, is often an imperfect fit. Chapter 9 (“Juveniles and Homicide”) and Chapter 10 (“Sexual Homicide”) are revelatory. These topics receive much publicity in the media, stirring up heated emotion but shedding little light. Why does an adult child molester sexually abuse a child and then bury her alive with her teddy bear? Why Columbine? Killers, their victims, and lethal circumstances are all highly individualized and multifaceted. Dr. Malmquist presents a balanced, detailed, in-depth psychological analysis of cases that richly inform the reader. We are quickly disabused of any stock answers regarding homicide. This one-of-a kind book will educate and intrigue both laypersons and professionals who want to learn more about homicide—a fit and proper quest for knowledge, not titillation. The final chapter, “Homicide in the Twenty-First Century,” acknowledges that we must know more about homicide. Are there genetic factors operating in murderers? If so, which genes, and how do they first become manifest? The answer would allow for early identification and intervention. Unfortunately, homicides will continue in this century and for centuries to come. It is our inescapable destiny. Dr. Malmquist has written a wonderfully comprehensive and insightful book on homicide that sheds much-needed light on the darker side of humanity. It will undoubtedly help save lives among the living—and those yet to come. Robert I. Simon, M.D. Clinical Professor of Psychiatry Director, Program in Psychiatry and Law Georgetown University School of Medicine, Washington, D.C. Chairman, Department of Psychiatry Suburban Hospital, Bethesda, Maryland

PREFACE TO THE SECOND EDITION

OVER THE COURSE of ten years since the first edition of this book was published, the topic of homicidal behavior remains a major problem in American society. Debate continues about its causes and cures both on a societal level and within the individual. These questions of causes and cures are part of what initially enticed me as a psychiatrist to pursue the study of individuals who carry out homicidal behavior. From these initial questions, the study of homicide has since progressed in an effort to place these individuals “under the microscope” utilizing whatever psychiatric understanding is possessed. The goal is not to bypass the knowledge existing in other fields, such as the social sciences and law, but to add to it. During the time elapsed since the first edition, psychiatrists have become more involved with the problem of homicide. Some of this involvement includes being called on to carry out assessments of homicidal individuals in legal settings and when a courtroom trial looms. The request may also be to provide further understanding of why such behavior occurs beyond the generally available demographic information. Involvement in homicide cases in the forensic area has expanded with legal issues arising at many points: from the accused perpetrator’s point of arrest, while the accused perpetrator awaits trial, during resolution of competency issues regarding proceeding to trial and other decisions that may arise (such as the perpetrator pleading insanity, refusing medications, or arguing his or her own case without a lawyer), and at the level of sentencing. The last item also puts psychiatrists into a testimonial role regarding execution possibilities in those states with capital punishment. At each of these levels, careful assessment of the types of personalities and their functioning is necessary. All too often, I have found that the “homicidal cases” have been viewed as homogeneous entities within the legal system. The legal fo-

ix

x


cus then became a matter of classifying the cases into a murder or manslaughter category of a certain degree. An alternative I pursued was to elaborate on the diverse types of personalities and disorders that are possible with a person who commits a homicide, utilizing a psychopathological perspective. Although the large amount of criminological and legal literature on homicide often provided insights from demographic factors and the legal questions raised, as a psychiatrist I felt that something was still lacking from the data to form a fuller picture. What I also realized was the need to address my own disturbing countertransference feelings that were elicited by those who had committed some grisly homicidal acts, such as those involving torture or mutilation. This countertransference became part of my motivation to understand the perpetrators. The overall number and rate of homicides in the United States annually has decreased, and there is concern about whether these numbers and rates will rise again. In the meantime, theories on why the overall numbers of homicides have decreased produce competing explanations, which are discussed in Chapter 1 (“Epidemiological Aspects of Homicide”). Certain topics in homicide have come to the fore over the past ten years. Specificity about different types of homicide is where increased attention will be focused in the future. The following are all topics that I realized needed discussion in an updated volume: • I found myself groping for more knowledge and possible explanations for why one particular type of homicide occurred in contrast to another type. The problem of serial homicides is one such area, and many questions abound. For example, why would a particular individual engage in an assault that leads to a homicide whereas another would engage in a vastly different type of behavior, such as a sexual homicide (Chapter 10, “Sexual Homicide”)? Knowledge has increased about biological and genetic factors operating in serious acts of violence and will continue to do so (Chapter 2, “Biological Factors in Homicide”). Both serial killings and sexual homicides have received increased attention and reveal gaps in our knowledge. There is the lurking question of whether rates of these killings have increased or whether we are just getting better at detecting these homicides. • Postpartum homicides of infants elicit intense passions about the legal blameworthiness of the parent (usually the mother), especially when much publicity surrounds a case (Chapter 8, “The Depressed or Bipolar Person and Homicide”). This is another area being explored in greater depth. The legal issues are only part of the scenario

Preface to the Second Edition

xi

because questions continue to arise about why such acts occur and whether they could have been prevented. • Homicides by those working in the health care fields have stirred an interest that has been neglected until a short time ago (Chapter 7,”Masochism and Homicide”). These homicides are now a matter of increased public concern, and again the question is whether they are increasing or just being detected more accurately. • Although the number and rate of homicides in the United States annually has decreased, concern has increased about the number of youths and juveniles who perpetrate a homicide (Chapter 12, “Homicide in the Twenty-First Century”). These homicides have occurred in contexts varying from gangs, random shootings, and within family settings and have raised questions about peers who have been informed beforehand about such possibilities (Chapter 9, “Juveniles and Homicide”). More recently, a complicating factor has arisen about whether selective serotonin reuptake inhibitors (SSRIs) increase the risk of suicide and/or homicidal violence for children and adolescents, and this question is under continuing investigation (Chapter 8, “The Depressed or Bipolar Person and Homicide”). Why young individuals continue to carry out homicidal acts raises broad questions not only about these individuals’ personalities, but the structure of families and our society at this time. One of the personal issues in writing a volume on homicide is deciding which cases to use for illustration and which to exclude. The temptation is to use cases that have gained wide publicity because we have all heard about them to some degree. The problem with using these cases are the limits on the degree and depth of knowledge available, especially from a clinical vantage point. The deciding principle used for this book was to use only case material where there was a legal conviction of guilt, a confession, or adequate clinical material to allow going beyond media-level reportage. Thus, an intriguing case, such as that of O.J. Simpson, is not discussed—not only because of a not guilty–verdict, but also because of an absence of clinical material we would need to elaborate on a man who was charged with killing his former wife, Nicole, and her friend, Ronald Goldman. To use popular cases from the media not only involves speculation, but also raises legal and ethical questions. Such concerns would hold similarly for other high publicity cases, such as that of Robert Blake (“Baretta”) who was found not guilty of murdering his wife. Another provocative, high-publicity case has been that of JonBenet Ramsey, a young girl murdered in 1996. The parents have been suspects, but no one has yet been charged with the

xii


crime. In the meantime, the parents wrote a book about the investigation, which they believe was bungled. These types of cases continue to be discussed but do not lend themselves to inclusion in a book dealing with the psychiatric aspects of homicide for the reasons noted. In contrast to the above types of high-publicity cases, postpartum cases, such as that of Andrea Yates, have presented adequate clinical material in the context of an insanity defense trial and commentary. The focus throughout this book continues to be the epidemiological and diverse clinical aspects of homicide where data is available and can be discussed from these perspectives. I have also been aware of the great expansion of the work of other mental health professionals, besides psychiatrists, who have become involved with homicide cases. The pure numbers of those in these fields have expanded faster than the numbers of psychiatrists, let alone forensic psychiatrists. The economic scene has changed in the health care field, in which increasing numbers of psychiatrists, psychologists, and social workers have seen work in the forensic area as an opportunity to try to maintain a private practice within a health maintenance organization world. This effort often has involved them in assessment and forensic work with homicidal individuals with whom they have had minimal experience or training, and I realized that many of them were relying on my book when confronting such cases. It has been personally gratifying to hear from those working in these different professions about the helpfulness of the book. Many have told me that such material either was not covered during their training or was covered only quite superficially. The result was that when readers came to the point of needing to assess firsthand a homicide case, prepare a report with an explanation, and then prepare to testify about someone who had perpetrated a homicide, using my book as a reference was most helpful. In a different vein, it has also been rewarding to be aware of how helpful the book has also been to those in the legal profession. Practicing attorneys in the private and public spheres, as well as judges, have mentioned its helpfulness in supplementing the legal analysis of a particular case and the issues raised. Similarly, the book has found a place in various law school courses or continuing educational seminars for attorneys, as well as those in the correctional field, who often find themselves working over many years with people who have committed a homicide. In courtroom proceedings, attorneys and judges have raised examination questions directly from the book. The hope is that this new edition will continue to be helpful not only to forensic psychiatrists, but to the general psychiatrists whose work

Preface to the Second Edition

xiii

becomes involved with people who commit, or raise a threat, of homicidal-level violence. This hope extends to other mental health professionals and those working in the legal setting. My interest in homicide and its perpetrators has come a long way since an esteemed professor and supervisor in my psychiatric residency days sincerely told me that working in the area of psychiatry and law was something I should postpone until retirement, because that is what retired psychiatrists often did. Perhaps that reality was once true, but it has certainly long passed.


PREFACE TO THE FIRST EDITION

MANY YEARS AGO, on first becoming involved in forensic psychiatric work and training, I became aware of the paucity of psychiatric materials dealing with the subject of homicide. To be sure, there were several insightful articles dealing with a particular kind or case of homicide, and a greater abundance of them dealing with the subject of insanity, but they did not offer an overall psychiatric perspective on homicidal behavior. Other disciplines, such as sociology and law, had an enormous bibliography of the subject with their own orientations. Bookstores also seemed perpetually filled with volumes dealing with popular crimes or murder mysteries. Yet these never seemed as intriguing as the actual cases I had personally examined on consultation. My approach was to determine whether any psychiatric problems existed and secondarily to become involved in reasoning about the complex legal issues presented. Several things pushed this situation toward a book. During a conversation with my old friend, Jay Katz, M.D., of the Yale Law School faculty, it came out that I had personally evaluated over 500 individuals who had committed a homicide. He urged that I share such experiences, and not merely keep them for my own curriculum vitae or my classes. The next step was a conversation with my colleague, David Ward, Ph.D., a criminologist and chair of the sociology department at the University of Minnesota. We decided to create and teach a course entitled “Killing.” It has been a popular course, and at first it seemed a logical step simply to create an anthology of articles on homicide. However, an anthology lacked a broader coverage of the topic that was based on clinical knowledge that one person could synthesize. It meant raising the types of questions that clinical and research psychiatrists would ask. This necessarily led to questions about the epidemiology of

xv

xvi


homicide, different diagnoses, descriptive diagnoses from the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, and how to integrate certain psychodynamic formulations for understanding and explanation. All of the above corresponded to the period of increasing concern about serious violence in the United States. Different people from backgrounds, not only in psychiatry but in psychology, sociology, and law as well, wondered if there was a psychiatric textbook available that could give them some overall insights on how psychiatrists thought about the problem of homicide. I began to realize that there was indeed an audience for such a work. Only a few colleagues indicated that this was a subject outside the province of psychiatry proper. In my thinking, homicide does not seem any more outside the province of psychiatry than suicide, for which psychiatry has never hesitated in accepting the possibility of psychiatric contributions. The subject of murder, in fact and fiction, is not likely to go away. In fact, the boundaries between real murder and novelized murder have become increasingly blurred in the media, such as the efforts to have some executions televised or videotaped for replay. My hope is that this book will shorten the search for other clinicians who have homicide cases, as well as attorneys, and that it will lead to an increase in articles and books in the field of homicidology. In terms of ideas and fruitful interchanges, it would be impossible to list all the people from whom I have benefited through discussions over the years. Besides Jay Katz and David Ward there have been Paul Meehl, Alan Stone, and Yale Kamisar at the University of Michigan Law School, who introduced me to the intricacies of criminal law, and diverse philosophers have all been stimulating. I also express my appreciation to Kate Stuckert for her patience and assistance in the typing of this manuscript in its several forms.

ACKNOWLEDGMENTS .. .there ain’t nothing more to write about, and I am rotten glad of it, because if I’d a knowed what a trouble it was to make a book I wouldn’t a tackled it, and ain’t a-going to no more.

—Mark Twain, The Adventures of Huckleberry Finn The group of people who initially captivated my interest in homicide— and who continues to do so—should first be given my appreciation. This group is composed of individuals who committed homicide and were seen by me in their roles as defendants or patients. They allowed me to try to understand what had happened in their mental life histories that had led to an outcome of homicide. Although homicide victims and their families are forced to confront the terrible consequences of this crime, the individuals who commit homicide, as well as their own families, also bear the heavy consequences of their actions and thus should not be slighted by clinicians. I continue to be grateful to many of my old mentors, professors, and friends who gave me the incentive to write this book. Among them are Andrew Watson, M.D., from the University of Michigan, now deceased; Jay Katz, M.D., from the Yale Law School; and Alan Stone, M.D., from the Harvard Law School. My continued interaction with my coteacher, Professor Eric Janus, J.D., of the William Mitchell College of Law in Minnesota, has always been fruitful and collegial. My old friend and colleague, Paul Meehl, Ph.D., now deceased, continues to influence my thinking because of his writings, critical mind, and dedication to truthfulness. Students in diverse fields have always provided a catalyst to my work by asking the basic questions that somehow elude us when we become too immersed in a particular subject. I appreciate the opportunity to teach these students from diverse fields over the years, among them psychiatric residents, graduate students, and law students. Without their questions, many areas in my study of homicide would have remained unexplored. Finally, I express my appreciation to Robert E. Hales, M.D., M.B.A., editor-in-chief of American Psychiatric Publishing, Inc., and John McDuffie, editorial director, with whom working has been a pleasant

xvii

xviii


and rewarding experience. In my department at the University of Minnesota, my personal thanks for technical assistance with parts of the book go to Karl Krohn, Ph.D., and Ms. Yoonie Helbig for secretarial assistance.

CHAPTER

1 EPIDEMIOLOGICAL ASPECTS OF HOMICIDE HISTORICAL PERSPECTIVE ON RATES OF HOMICIDE Trends in deaths from homicides for the twentieth century reveal low rates early in the century, with a peak first reached during the economic depression years of the 1930s. 1 The homicide rate dropped during World War II, followed by a slight increase in the late 1940s and early 1950s. By the end of that decade, the rate had dropped, but it picked up again in the 1960s with a steady climb. The peak rate of 10.2 murders per 100,000 people was reached in 1980. During the 1980s, the homicide rate continued to be among the highest in the twentieth century in the United States. From 1990 to 1994 the rates remained quite high, ranging from 9.0 to 9.8, but then began to drop. In the last years of the century, the rates for serious felonies, including homicide, began to drop and continued to drop into the twenty-first century, with a rate of 5.5 in 2004. Such data are available from the annual publication of the Uniform Crime Reporting (UCR) Program by way of reports to the Federal Bureau of Investigation (FBI) and its Supplementary Homicide Reports (SHRs) and are summarized in Table 1–1. The debate has been why these changes occurred. In a report by the National Academy of Sciences,2 it was noted that the annual risk of becoming a homicide victim had risen from 1 in 12,000 in 1987–1988 to 1 in 10,600 by 1990. However, the lifetime risk of homicide was much greater, with the highest lifetime rate in six demographic groups being among black males. Their rate of 4.16 is equivalent to a 1 in 24.1 chance of dying by homicide. For American Indians, the risk is 1 in 57, whereas for white males and females the risk is less than 1 in 100. It must be emphasized that despite the media’s focus on killings of adolescents and young males, less than 25% of one’s lifetime homicide risk is actually incurred by the time that person reaches his or her 25th birthday. Thus the high number of black males dying by age 24 years must be seen in the context of the high homicide rate for black

1


2 TABLE 1–1.

Murder and nonnegligent manslaughter in the United States

Year

Population

Number of reported offenses

Rate per 100,000 inhabitants

1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

231,664,458 233,791,994 235,824,902 237,923,795 240,132,887 242,288,918 244,498,982 246,819,230 249,464,396 252,153,092 255,029,699 257,782,608 260,327,021 262,803,276 265,228,572 267,783,607 270,248,003 272,690,813 281,421,906 285,317,559 288,368,698 290,800,000

21,010 19,308 18,692 18,976 20,613 20,096 20,675 21,500 23,438 24,703 23,760 24,526 23,326 21,606 19,645 18,208 16,974 15,522 15,586 16,037 16,204 16,503

9.1 8.3 7.9 8.0 8.6 8.3 8.5 8.7 9.4 9.8 9.3 9.5 9.0 8.2 7.4 6.8 6.3 5.7 5.5 5.6 5.6 5.7

Source. U.S. Department of Justice: Crime in the United States. Washington, DC, U.S. Department of Justice, 2004.

males at all ages. Arrest data for 2003 indicate that males composed 89.7% of murder arrestees. By race, whites accounted for 49.1% of the murder arrestees and blacks 48.5%, even though only 12% of the population is black. Data from the UCR indicate that a record number of 23,438 overall homicides were committed in 1990.3 However, this figure by itself is relatively meaningless; it must be viewed in relationship to other timemeasured variables to have any meaning. For example, an increase in population could lead to an increase in the number of homicides. This factor has led demographers to study longitudinal trends that include homicide rates. The overall rates can vary significantly over time. In addition, the data may be broken down into rates for various states and regions of the United States and may detail the relationships between the people involved in the killing (e.g., homicide victims being spouses,

Epidemiological Aspects of Homicide

3

children, and so forth). For example, in the early 1990s, the western United States appeared to supplant the South as the region with the highest overall level of violence among the four regions the UCR program uses. However, by 2003 the southern states, which make up 35.9% of the nation’s population, had regained the position by accounting for 43.6% of the murders in the country.4 Another important variable is the demographics, such as the proportion of those in a high-risk age group for committing homicide at a particular time.

SPECIFIC EPIDEMIOLOGICAL ASPECTS Nine key epidemiological variables have been studied or raised as significant with respect to homicide: age, race, gender, socioeconomic class, method of killing, relationship between the victim and perpetrator (including cases that some interpret as being precipitated by the victim), the perpetrator’s prior arrests, use of alcohol or other drugs, and temporal and ecological factors. One caveat is needed before exploring the individual epidemiological variables. Epidemiological data reflect data on the victims but not on the perpetrators. This omission is interesting in its own right and perhaps signifies that understanding the mind of the murderer requires a clinical or criminological explanation. Thus, in an operational sense, homicide rates can only give the numbers of people killed in any given population unit. Many of the factors discussed in this section overlap, and no scientific presentation can make a claim that any one variable is the cause of a homicide. Instead, a workable model for determining factors leading up to a homicide must elicit significant background factors that under certain precipitating circumstances, predispose an individual to commit a homicide. The inductive search for determining factors must be an empirical one in which the basic premise is that for any one homicide, several factors are necessary but one factor is never a sufficient cause for that homicide. Such a position is compatible with considering the significance of personality attributes and given clinical diagnoses. It is also consistent with the position that when legal questions arise regarding a person’s responsibility for a homicide, no answer can ever be given simply on the basis of a diagnosis. Overdetermination of homicidal actions is the rule.

Age Much publicity has focused on homicides committed by juveniles from the early 1990s onward, when high rates were disproportionately contributed

4


by young black males. This accompanied the concern about a drug culture in which homicides might be a part of the trade. A concealed factor then and now is that juvenile homicides are often conducted by more than one party but with only one victim, which skews juvenile offenders in the statistical distribution. Even taking the past 25 years through 2002, only 10% of murderers and victims were under 18 years of age, whereas the 18- to 24year-old group contained about 36% of offenders and 24% of victims.5 Tables 1–2 and 1–3 illustrate the number of murder victims and offenders by age, sex, and race for 2003. Note the relatively high victim rates for infants, which then decrease, with the lowest rate found among those ages 5–12 years. White female infants have had the greatest risk of all white females for dying by homicide, a risk related to acts of infanticide. This topic is discussed further in Chapter 8 (“The Depressed or Bipolar Person and Homicide”). The loading of males as murder offenders is significant in the age groups of 17–19 years and 20–24 years, with especially high numbers of black males distributed in these groups.

Race The racial factor is reflected in fluctuations over time in general, but again, rates for nonwhite homicide victims and perpetrators are higher than for white victims and perpetrators for every age group. Depending on the age group studied, the rate for homicides among nonwhite males can be as high as 10 times that for white males. It should be noted that most studies have focused on larger American cities. In a study of homicide in New York City, Tardiff and colleagues6 found that young black and Latino men were more likely to be homicide victims than men from any other demographic group. In the subset who died within 48 hours, 31% were positive for cocaine metabolites, and 25% of their deaths involved firearms, which suggests a basis for the high death rate. In turn, these variables may be situational and related to poverty. The significance of race in epidemiological studies of homicide cannot be ignored. Yet the deeper question is, what does it mean? Besides the national data noted earlier, other studies have documented that black people are involved at a rate far in excess of their numbers in the population. The studies are also consistent in indicating that such killings are overwhelmingly intraracial. Findings from Wolfgang’s7 early study in Philadelphia, PA, reported that based on 1948–1952 data, 73% of the offenders and 75% of the victims of 588 homicides were black; these findings were consistent with a later study by Hewitt.8 Furthermore, the distribution seems to be continuing even though the overall murder and manslaughter rates have declined from the early 1990s.

Murder victims by age, sex, and race, 2003 Sex

Age (years) Total Distribution (%) Under 18 Under 22 18 and over Infant (under 1) 1 to 4 5 to 8 9 to 12 13 to 16 17 to 19 20 to 24 25 to 29 30 to 34 35 to 39 40 to 44 45 to 49 50 to 54 55 to 59 60 to 64

Total

Male

14,408 100.0 1,333 3,445 12,811 225 307 82 69 369 1,283 2,855 2,148 1,594 1,286 1,114 951 630 365 226

11,167 77.5 905 2,724 10,083 131 164 41 36 293 1,113 2,432 1,826 1,249 973 783 694 463 256 160

Female 3,215 22.3 424 716 2,721 91 142 41 33 76 169 420 322 344 313 330 256 167 109 66

Race Unknown 26 0.2 4 5 7 3 1 0 0 0 1 3 0 1 0 1 1 0 0 0

White

Black

6,913 48.0 660 1,556 6,133 139 165 44 33 150 549 1,165 839 692 636 589 552 382 236 148

6,887 47.8 617 1,749 6,189 73 131 34 34 202 689 1,585 1,234 842 605 484 362 226 113 57

Other 408 2.8 40 100 358 7 6 3 1 14 34 74 54 43 36 32 26 19 12 16

Unknown 200 1.4 16 40 131 6 5 1 1 3 11 31 21 17 9 9 11 3 4 5


TABLE 1–2.

5

6

TABLE 1–2.

Murder victims by age, sex, and race, 2003 (continued) Sex

Age (years)

Total

Male

65 to 69 70 to 74 75 and over Unknown

164 153 323 264

113 90 171 179

Female 51 63 152 70

Race Unknown 0 0 0 15

White

Black

Other

122 105 247 120

32 42 61 81

8 4 9 10

Unknown 2 2 6 53



Murder offenders by age, sex, and race, 2003 Sex

Age (years)

Total

Male

Female

Unknown

16,043 100.0 813 3,327 9,396 0 0 1 12 432 1,585 2,780 1,641 1,054 800 713 471 277 166 94

10,218 63.7 726 3,049 8,387 0 0 1 9 381 1,470 2,538 1,483 921 672 598 407 242 141 87

1,123 7.0 86 275 997 0 0 0 2 51 114 240 157 129 128 113 63 34 25 7

4,702 29.3 1 3 12 0 0 0 1 0 1 2 1 4 0 2 1 1 0 0

White 5,132 32.0 351 1,366 4,489 0 0 1 7 185 625 1,147 696 533 431 448 266 174 114 71

Black

Other

Unknown

5,729 35.7 433 1,832 4,542 0 0 0 4 229 889 1,542 885 480 340 235 185 91 43 19

308 1.9 21 95 280 0 0 0 0 13 49 76 45 30 23 23 17 9 7 3

4,874 30.4 8 34 85 0 0 0 1 5 22 15 15 11 6 7 3 3 2 1

7

Total Distribution (%) Under 18 Under 22 18 and over Infant (under 1) 1 to 4 5 to 8 9 to 12 13 to 16 17 to 19 20 to 24 25 to 29 30 to 34 35 to 39 40 to 44 45 to 49 50 to 54 55 to 59 60 to 64

Race


TABLE 1–3.

8

TABLE 1–3.

Murder offenders by age, sex, and race, 2003 (continued) Sex

Age (years)

Total

Male

65 to 69 70 to 74 75 and over Unknown

66 44 73 5,834

60 43 60 1,105

Female 6 1 13 40

Race Unknown 0 0 0 4,689

White 48 33 61 292

Black

Other

16 9 8 754

1 2 3 7

Unknown 1 0 1 4,781




9

These data are depressing and unremitting, and they cry out for an explanation as well as an amelioration of the situation. Young black males and their families take the brunt of this public health problem. When it is realized that blacks account for only about 12% of the U.S. population but disproportionately account for approximately 50% of those arrested for murder and 35% arrested for assaults, obviously there is a problem that has so far remained unsolved. In terms of rates, blacks are six times more likely to be victimized and eight times more likely to commit homicides than whites. Figures 1–1 and 1–2 illustrate the different rates of victimization and offending of blacks and whites for homicide in 2002. It can also be noted that in terms of the years of potential life lost, homicide exerts a greater effect than many other causes of death, including suicide, because homicide victims are from a younger group on average. This in itself would make it a major social problem. Several explanations have been offered for these findings. Traditional explanations have raised the possibility of social and economic inequality, lack of opportunity, racism, and discriminatory practices in the criminal justice system, ranging from the point of arrest to sentencing. Some have argued that such disparities exist for minorities in diverse cultural settings outside the United States as well. Yet although all these factors are relevant, in themselves they do not dispose of the question. To complicate matters, the high-risk group of black males—those in the 14–24 age group—has changed very little in terms of numbers from 1993 to the present overall decline in homicide rates. It is interesting to contrast these data with scholarly studies of why long-term homicide rates have declined in Europe. Explanations are offered in terms of a decline in elite homicide (i.e., homicide by the upper social classes) over a long period and a drop in male-to-male conflicts in public space, but the racial factor is omitted, although noted age and sex patterns remain relatively unchanged.9 Consider the explanation of economic and social deprivation. Both topics are raised as a correlate of homicide that can have some possible causal significance or possibly none at all. In consideration of how these factors may operate, the concepts of poverty as a state of absolute deprivation and inequality as a relative deprivation are relevant. The result is that fewer economic resources are available for those in such a predicament. If the focus is on the relationship between poverty and homicide rates in large American cities, there is a consistent and positive correlation to all homicide rates. What is important is that this correlation exists independently of variables such as race, region, or population size. However, various kinds of inequality do not show an effect on homicide rates, also independent of these same variables.10

Homicide rate (per 100,000)

10

White Black Other

FIGURE 1–1.

Homicide victimization by race, 1976–2002. Rate per 100,000.

Source. Bureau of Justice Statistics: Homicide Trends in the United States. Washington, DC, U.S. Department of Justice, 2004.


Year



White Black Other

Year

FIGURE 1–2.

Homicide offending by race, 1976–2002. Rate per 100,000.


11


12

Gender Data from the Bureau of Justice Statistics of the U.S. Department of Justice for a 25-year period ending in 2000 indicated that for all homicides, males were the offenders in 87.9% of cases and victims in 76.4%; females were the offenders in 12.1% of cases and victims in 23.6%. Women were especially at risk for intimate killings, sex-related homicides, and murder by arson or poison. In turn, women were more likely to commit murder as a result of an argument or to murder by poison. These rates can be further broken down as follows: Male offender/Male victim, 65.2% Male offender/Female victim, 25.0% Female offender/Male victim, 7.2% Female offender/Female victim, 2.6% Some of the statistical findings regarding gender can be summarized as follows: 1. At the beginning of the twenty-first century, males were 3.2 times more likely than females to be murdered. 2. Males were 10 times more likely than females to commit murder. 3. Males were more likely to be victim targets of homicides than females. 4. Rates for females as homicide offenders have declined since the early 1980s, whereas rates for males as offenders peaked in the early 1990s and then began to decline. 5. Most males murder men, but most females who murder also murder men. 6. If anything, the gender gap between male and female murderers has seemed to widen and not narrow as we have moved into the twentyfirst century. Regarding perpetrators, in 80% of the cases, men killed nonintimate acquaintances, strangers, or victims of undetermined relationship, whereas women killed nonintimately related people in 40% of cases. When men killed with a gun, they most commonly shot a stranger or nonfamily acquaintance. When women killed with a gun, the victim was five times more likely to be a spouse, family member, or intimate acquaintance. Perhaps this finding reflects men being more conversant with the use of guns as well as more predisposed to use them in arguments. Women prefer a method of killing by such means as arson or


13

poisoning and for infanticides, suffocation. Even less common are the situations in which a woman hires someone to murder her lover or spouse, although the scenario makes for good television drama. When the patterns for race and gender are examined, an interesting variation emerges. Rates for men in the white and nonwhite categories are higher than for women in the corresponding racial categories. Rates for nonwhite individuals are higher than those for white individuals, partially based on the fact that homicide rates for nonwhite women are greater than those for white men. Separating the effects of race from those of gender may be difficult and is complicated by the frequent connection between gender and killing. In Wolfgang’s early study,7 82% of the perpetrators of killings and 76% of the victims were male. The race factor is also strongly connected to gender, as witnessed in the homicide offender rates being 41.7 for black males, 7.3 for black females, 3.4 for white males, and 0.4 for white females. From year to year, the UCRs do not show much annual variation in the arrest rates for homicide by gender, which is approximately 90% for males and 10% for females. The 2003 data from the UCR indicate that of the 16,503 homicides, 89.7% of arrestees were male. However, as noted, such statistics cloak differences in the types of homicidal acts committed as well as the clinical differences in such acts. Bartol11 noted that the 9:1 ratio is consistently found in studies for many major American cities as well as for some cities in England and Israel. However, some intriguing cultural differences do emerge. In England, a far higher number of the victims are females (60%) than in the United States (25%). In an Israeli study12 on Jewish homicides, 51% of Jewish victims were female, compared with 34% who were non-Jewish. However, in absolute numbers, these homicides are still fewer in number than in the United States, presumably because of the effects of alcohol, drugs, and crimes related to drugs (e.g., robberies).

Method of Killing As to the circumstances of killing, 2003 data from the UCR indicate that most homicides are committed with a firearm, with the largest subcategory of firearm used being a handgun in 53% of those homicides. Firearms are involved in about 67% of homicides in the United States.13 In addition, some states estimate 5.7 nonfatal gunshot injuries for every homicide.14 However, the risk of death by firearms may be skewed by a particularly high elevation among black male adolescents. Earlier studies, when the homicide rates were peaking, revealed the ratio of indi-

14


viduals ages 15–19 years who were victims of gun-related homicide was 7.5 per 100,000 for white males but 83.4 per 100,000 for black males. These data were believed to be correlated with death by firearms, in which the fraction of homicides committed with guns peaked in those ages 15–19 years, with 81% becoming victims by that means.2 The skewing is reflected in the percentage of homicide victims killed with a gun, which increases with age up to age 17 and then begins to decline. The UCR data for 2003 also indicate that knives or other cutting instruments are used next frequently, followed by blunt objects or parts of the body (e.g., hands and feet used as weapons or clubs and hammers). At the bottom of the frequency scale are more unusual measures, such as poison, drowning, strangulation, asphyxiation, arson, or explosives. A comparison of the risk of death versus a nonfatal injury during family and intimate assaults reveals that the use of a firearm is 12 times more likely to result in a death than non-firearm-associated events.15 Therefore, firearms play a significant role not only in homicides committed by men but also in the context of family and intimate assaults where there is access to such a lethal weapon. Another component of the circumstances of the homicide deals with the setting of the homicide. Some of these settings are associated with murders connected to felonies or lesser crimes. The SHRs try to provide details about specific homicides, such as romantic triangles, brawls under the influence of alcohol or narcotics, various types of arguments often unspecified, gang killings, juvenile gang killings, and so on. For the homicides committed in 2003, the circumstances surrounding the killings were actually unknown in 33.9% of cases. The largest identified group of 16.4% was connected to a felony murder, such as during the commission of a robbery or narcotic violation. Among the heterogeneous group not connected to the commission of a felony, arguments were cited as behind 28.6% of the killings. These data provide little specificity in terms of what occurred during the argument between the participants or a group but simply classify the setting of the homicide to the best extent possible.

RELATIONSHIP BETWEEN VICTIM AND PERPETRATOR Although the victim–perpetrator relationship in homicides is often one of family ties or intimate relationships, the number of homicides committed by strangers has increased significantly in recent years. As characterized in the UCRs, the majority of homicides occur in the context of what is referred to as primary homicides, with primary referring to a death

15


that occurs in a relatively spontaneous fight between people who know each other, often over money or property and fueled by alcohol or drugs. According to the 2004 UCR, 12.5% of perpetrators and victims were members of the same family and another 30.5% were friends, neighbors, and acquaintances. Only 12.5% of homicides occurred among strangers. However, in 44.5% of these cases, the police could not determine whether a prior relationship existed between the perpetrator and the victim. This may be because they cannot discover who the perpetrator was or because there is a lack of evidence to make any connection. There is also a gender difference in victim–offender relationships, in which female victims are more likely to be killed by an intimate or family member. In contrast, male victims are more likely than female victims to be killed by acquaintances or strangers (Table 1–4). If the stranger and undetermined relationship categories are combined, these ratios amount to 51.8% for males and 36.6% for females.

TABLE 1–4.

Victim–offender relationships by victim gender, 1976–2002 Homicide victims by gender,%

Victim–offender relationship

Male

Female

Total Intimate Spouse Ex-spouse Boyfriend/Girlfriend Other family Parent Child Sibling Other family Acquaintance/Known Neighbor Employee/Employer Friend/Acquaintance Other known Stranger Undetermined

100.0 5.5 3.3 0.2 2.0 6.7 1.3 2.1 1.2 2.1 35.9 1.2 0.2 30.1 4.5 15.5 36.3

100.0 30.1 18.6 1.4 10.1 11.4 2.7 5.4 0.9 2.5 21.9 1.3 0.1 17.2 3.2 8.9 27.7

Note. The victims of the September 11, 2001, terrorist attacks are not included in this analysis. Source. Bureau of Justice Statistics: Homicide Trends in the United States. Washington, DC, U.S. Department of Justice, 2004.

16


It is thus probable that more people are being killed by strangers than ever before. Data from the Bureau of Justice Statistics indicate that the number of homicides in which the circumstances are unknown has tripled since 1976.16 On the other hand, homicides involving adult or juvenile gang violence increased more than sixfold in that time period. In the past, there has been a customary 70%–75% “clearance rate” for homicides, where the police believe they have solved the crime and one or more persons should be arrested. The clearance has been associated with a high rate of reporting, because few homicides are never reported to authorities. By 2003, the clearance rate had decreased to 62.4%, presumably related to the increase in stranger killing. It appears that about 50% of homicides stem from an argument, whereas another 25% occur under the influence of alcohol or drugs. This discussion leads to the broader topic of how an epidemiological approach conceptualizes perpetrator–victim relationships and their potential for heightening the possibility of a homicide. One approach has been to use a victim-precipitation model, which has been criticized as blaming the victim. An alternative approach has focused on the diverse and complex interactional variables in relationships or situations that can have a violent ending. The victim-precipitation model looks at the ways homicide victims played some role in their own future deaths, focusing on situations such as arguments in bars or disagreements between spouses and lovers that escalated into increasing taunts, insults, or physical contact. When the situations were later reconstructed from the available facts, it often seemed that what had pushed matters toward homicide often began with a trivial event or exchange. The presumption in such theorizing is that the individuals, although not necessarily intimates, were interacting with each other, and therefore the killing would not be seen as part of a model for killing by a stranger. This model would thus not be applicable to killings occurring in the context of robberies or burglaries. In an interesting approach that parallels many psychiatric theories of interpersonal violence, criminologists look at the types of interactions between individuals that eventually lead to a death. The difference between this theory and earlier theories about victim-precipitated homicide is the realization of how a homicidal outcome is a “situated” transaction that could easily have many possible outcomes. Frequently, a situation escalates into violence because neither participant allows the other the opportunity to back away while still saving face. These escalations are neither the stuff of dramatic television portrayals nor the act of a deranged killer who could later plead not guilty by reason of insanity. Instead, these types of killings arise in the give-and-take of ordinary transactions between people and comprise a frequently found framework in which homicide is enacted.


17

Miethe and colleagues17 conducted a study of situational contexts in homicides, focusing on the interactions among offenders, victims, and elements of the offenses. They relied on data from narrative accounts and the SHRs over three decades using a methodology of qualitative comparative analysis. One of their findings was that categorizing homicides as either instrumental (those carried out for an explicit future goal, such as acquiring money or status) or expressive (unplanned acts of anger, rage, or frustration) is excessively simple. There is so much variance among the nature and prevalence of the offender, victim, and situational factors that the distinction between instrumental and expressive motives is unclear. The configuration of both is required to begin to understand the social processes and triggering mechanisms involved when individuals and situations converge for a homicide. Psychiatry’s major contribution to comprehending how these seemingly commonplace behaviors lead to homicides is in terms of understanding personality vulnerability and functioning and the impact of painful emotions on human beings. These painful emotions are evoked in the course of interactions in which sensitivities or misinterpretations are played out. In some situations, certain personality disorders or traits predispose the individual to overreact or misinterpret. For these people, the sway of emotions then surges to the surface and erupts into some impulsive action. The actors might be spouses or lovers who argue and then progress to shoving and slapping and then to the use of fists, objects, or weapons. Name-calling or taunts are frequent. In some cases, it might be a child’s disobedience or the endless crying of an infant that elicits feelings of helplessness in the adult, who then reacts to the child’s behavior as a provocative challenge that must be handled. At a crucial juncture, the point of no return is crossed. Luckenbill,18 a criminologist, broke down these various types of interactions into a series of stages. In stage one, the future victim performs some type of act or verbalization that is interpreted as an attack on the future perpetrator’s activities or self-esteem. Seen from a psychiatric perspective, this initial stage would reflect the perpetrator’s tendency to distort or overinterpret. For example, an individual with paranoid or borderline personality would be among those likely to overreact. Stage two is an attempt to confirm the meaning of what the potential perpetrator has heard. Other people may be turned to for their interpretation; in fact, it is interesting that a majority of such confrontations occur in the presence of others. By the time stage three is arrived at, the perpetrator is at the point of responding, often by challenging the intended victim or demanding a halt to or retraction of certain statements. Threats may emerge. Stage four is the counterresponse of the intended victim, who at

18


that point does not back off but stands his or her ground. Stage five is the emergence of violence, either abruptly or after the perpetrator seeks out and obtains weapons. Stage six involves the reaction of the perpetrator after the killing, such as fleeing or being restrained by others. In this framework, many variables are operating, and indeed are necessary, for the predicted final outcome to occur; these variables involve both instrumental and expressive motivations. One necessary element is the occurrence of an intense interaction in which others are usually present but in which no one intervenes. The ready availability of weapons as a part of a high-risk lifestyle may also be an operating variable. Such a milieu is typically present in places such as bars, entertainment clubs, or street-corner confrontations. Note that through the first four stages of this model, various outcomes are possible.

PERPETRATORS’ PRIOR CRIMINAL OFFENSES There are many difficulties in assessing the significance of prior delinquent or criminal behavior with respect to the later commission of a homicide. A major problem is related to risk assessment or the prediction of such behavior. There is the further complication of moving from the realm of early adult criminal behavior to also encompass the antecedent role of adolescent antisocial behavior and whatever its relationship is to later violence in the form of becoming a future murderer. Specifically, much difficulty is attendant on predicting rare events such as homicides. When a retrospective look is taken after a homicide has occurred, certain variables often appear prominent. However, such retrospective approaches do not allow much sensitivity when it comes to antecedent prediction. A practical problem is that juvenile files are sealed in most states unless the juvenile has been waived to the adult criminal justice system. At that point, and for most young adults, it is usually not their first encounter with the legal system. Such data are not to be interpreted as indicating that most juvenile offenders go on to become violent adult offenders. This is not the case. The data indicate that of the 33%–45% of males in the United States who were once detained by the police or arrested as juveniles, the majority (approximately 65%) were arrested only once or twice and have no subsequent adult record. What these data mean is that a small number of juvenile offenders persist in committing crimes into young adulthood and that those juveniles who engaged in serious crime at an early age form a subgroup of individuals who are more likely to engage in serious violence, such as a homicide, in their 20s.


19

The highest rates for victimization and perpetration of a homicide occur among older teens and young adults. The rates for these groups increased dramatically in the late 1980s and early 1990s and then declined, as they did for all ages, as shown in Figures 1–3 and 1–4. In the period from 1985 to 1992, the homicide victimization rates for 14- to 17-year-olds increased almost 170% and then declined to levels similar to those experienced between 1976 and 1985. The highest victimization rates in that period were in the 18- to 24-year-old age group, which replaced the 25- to 34-year-old group. Offending rates of 14- to 17-year-olds increased rapidly after 1985, surpassing the rates of 25- to 34-year-olds, but the latter group has now recovered its higher position over the teenagers. The 18- to 24year-old age group has the highest homicide rate, and from 1985 to 1992 their rates almost doubled; the rates declined since then but still remain higher than their levels prior to the mid-1980s. At the height of these killing rates, the idea of a group of juveniles and young adults who were “superpredators” was broached. The implication was that this would be the group from which future killers would emerge. However, it was realized that many of these killings were not planned acts of homicide but rather impulsive acts often connected with drug dealing. They were thus often gang related and carried out by two or more individuals. Much of the impetus for the killing came from the combination of drug dealing and the easy availability of guns. The phenomenon of group identification and the idea of needing to join in group activities or losing a reputation was often at stake. There have been endless discussions about what contributed to the lowering of rates in this age group. Suggestions included the shift away from crack cocaine, police adopting a mixture of new tactics varying from cracking down to fostering more community involvement, and schools taking on monitoring roles such as checking for weapons. However, the rise and fall of gangs will inevitably continue, as witnessed by gang-related killings in large cities again increasing to a level double that of 1999.19 The early onset of delinquent behavior is often seen as a good predictor of later chronic, serious, and violent criminal acts when the youth becomes a teenager or young adult. When a juvenile’s early offenses include serious violent offenses (e.g., robbery), his or her probability of committing serious crime as an adult is increased. A third predictor variable in the juvenile offender is repetitive property offenses. Delinquency convictions before age 16 years and repeated sentences to juvenile correctional facilities are two other significant variables. When these variables are taken together in a weighted summative manner, the probability increases that a juvenile offender will commit a violent offense—including an increased possibility of committing a homicide—


20

Age in years Under 14 14–17 18–24 25–34 35–49 50+

FIGURE 1–3.

Homicide victimization by age, 1976–2002. Rate per 100,000.



Year


Under 14 14–17 18–24 25–34 35–49 50+


Age in years

Year

FIGURE 1–4.

Homicide offending by age, 1976–2002. Rate per 100,000. 21


22


as an adult. In Wolfgang’s study7 of 588 homicides, two-thirds of the perpetrators had prior arrest records. It was revealing that most of these arrests had not been for trivial offenses, nor even for property offenses, but for offenses against people that usually had been violent. Meta-analysis has been used to try to get at predictor variables in longitudinal studies of antisocial behavior as related to later violent behavior. Although statistical significance is not the same as practical significance in identifying who is at risk for violent behavior, it is known that only a small number of individuals will engage in homicide-level violence in adolescence or early adulthood. Meta-analysis allows small samples to overcome the low base-rate problem. By this approach, the strongest predictors for later serious violence beginning at ages 6–11 years were committing general offenses and engaging in substance use (chiefly tobacco and alcohol). A second-ranked group of predictor variables was male gender, family socioeconomic status, and antisocial parents. The strongest predictors at ages 12–14 years were lack of social ties, antisocial peers, and committing a general offense. A third-ranked set of variables was aggressive behavior, school attitudes and performance, various psychological conditions, parent–child relations, and physical violence. The conclusion for these groups is that these characteristics can distinguish juveniles who are at risk for violent, serious offending later in the 15- to 25-year age range.20 Of course, such data can never specify who is going to commit a homicide, but only the group who is at higher risk. The psychiatric significance of these data is that they indicate that homicide perpetrators had unresolved conflicts involving family problems and aggression and that there had been a failure to contain that aggression. It can be hypothesized that these conflicts are replayed in later acts of violence against individuals that culminate in a subsequent homicide. Although criminological researchers have not studied just what these psychological or social conflicts might be, this hypothesis points to the need for parallel lines of inquiry. It would also be significant to determine which intervening variables contributed to the remaining one-third in Wolfgang’s study7 later committing a homicide or whether these individuals were just “lucky” that earlier personal violence did not lead to arrests.

ROLE OF PSYCHOACTIVE SUBSTANCES Alcohol or other psychoactive drugs are mentioned so frequently in connection with various incidents of violence, including homicide, that the


23

specific role they may play is often not discussed. In fact, the words alcohol and violence are often used together as though they are inseparable, with the implication that alcohol causes the violence. This belief has become almost dogma among many in top political positions and has been reflected in legislative and community decisions involving fiscal distributions. The preconception is that alcohol facilitates the release of aggression. This position is usually based on data and figures that indicate a high percentage of violent acts are perpetrated by individuals under the influence of alcohol, even though in many cases the blood alcohol level of perpetrators was not taken, and the connection was made solely based on observation. The same thinking applies to other drug usage, in which, despite the varying effects of different drugs, the suggestion is that the person imbibed or took a drug shortly before a homicide occurred. Accurate tests to confirm or disconfirm the connection between homicidal violence and alcohol use would need to compare data between perpetrators who have ingested alcohol or psychoactive substances and individuals who have ingested similar substances but have not acted in this manner; these tests would also require control of variables such as when and where the drinking or drug use took place. It is easy to bypass other significant factors involving drug use besides the psychopharmacological effects. There is the resort to crime to get money to buy drugs, which increases the chance of violent encounters. In addition, the drug dealing trade may involve competition and intimidation techniques that can lead to homicides. When such dealings involve gangs, the potential for homicides is magnified. Drug abuse, especially alcohol, is also involved in many vehicular homicides. In the absence of any definitive studies of these factors, a large number of studies suggest a relationship between psychoactive substances and homicide on a biological or sociological level. In Holcomb and Anderson’s study21 of males charged with first-degree murder in Missouri, the authors found that 55% of the perpetrators had ingested drugs, alcohol, or both at the time of the killing. Mayfield22 found that 57% of convicted murderers studied had been drinking at the time of the killing. Greenberg 23 reported that alcoholism was diagnosed in 20%–40% of convicted murderers. A similar figure was noted in a California study of females arrested on homicide charges: 51% had been drinking at the time of the killing.24 Some 60% of violent offenders are estimated to have consumed alcohol at the time of their offenses.25 Roizen26 found that 86% of homicide offenders had been drinking at the time of the offense. Results from such studies support the generalization that about half of homicides occur under the influence of drinking. What the studies do not bring out sufficiently is that in many of these situations, the perpetrator and victim have both been drinking. In Wolf-

24


gang’s study,7 only the perpetrators had been drinking in 11% of the homicides, only the victims had been drinking in 9%, but both the perpetrators and victims had been drinking in 44%. Subsequent studies support a figure of close to 50% for the percentage of victims who had been drinking at the time of their deaths.27 However, despite the common assumption that alcohol, in amounts that are not sedative, increases one’s potential for aggression, the action– reaction relationship is not linear. The outcome depends on a host of variables, such as the cultural setting, genetic vulnerability, neuroendocrinological factors, other drugs that may have been taken, and immediate precipitants. Hence, the mechanism for how a violent end-point occurs is not entirely clear. Some studies find little evidence of a simple direct relationship between the psychoactive effects of drugs or alcohol and aggression.28 The pattern of drinking may play a role, such as in the more concentrated style of drinking prevalent in the United States rather than the spaced drinking style prevalent in some European countries. The suggestion of a type of alcohol-facilitated, episodic dyscontrol syndrome arises. Yet results from studies in Scotland29 and the Bordeaux region of France30 have indicated that 55% and 51%, respectively, of those convicted for murder had been drinking at the time of their offense. One persistent source of controversy is the extent to which someone carrying out a homicide under the influence of alcohol is responsible for that act. The answer to this question, in turn, is connected in part with a belief system, neither proven nor disproven, that alcoholism is a disease. The inference is that individuals who are biologically vulnerable to the effects of alcohol should not be seen as responsible for their violent acts committed under the influence. This type of reasoning has often been subjected to critical appraisal.31 In addition, not everyone subscribes to the theory that the loss of inhibition accounted for within a biological disease model of alcohol use is the crucial variable in assessing personal responsibility for behavior. Instead, proponents of a social-cognitive model have focused on the learned expectations of what alcohol supposedly permits one to do. Thus if the expectation is that when someone drinks, his or her control over personal impulses will be lost, then it is more likely that he or she will lose control. Bartol11 summarized the debate from this social-cognitive perspective, in which the consumer’s expectations of the effects of alcohol act as a crucial variable in how he or she will behave when drinking. Those holding to the social-cognitive model see the psychoactive influences of alcohol as learned and situationally determined and thus see the loss of control as related to a cognitive expectancy rather than a disease.


25

On the other hand, no one can doubt that alcohol has physiological effects on the nervous system. A correlation exists between low cerebrospinal fluid concentrations of 5-hydroxyindoleacetic acid (5-HIAA), a serotonin metabolite, and impulse control problems.32 Impulsivity and disinhibition have consistently been related to individuals with chronic alcoholism as well as related to Cluster B personality disorders.33 Also, the results of genetic work on alcoholism and antisocial personality have raised questions about whether there is some overlap among those who carry one or both diagnoses.34 There also appear to be differences between males and females regarding the connection between alcohol consumption and violent behavior that may be related to different social expectations.35 In this connection, there is an observation that men using cocaine are more likely to perpetrate violent crimes, whereas women using cocaine are more likely to become victims of violence.36 The debate concerning alcohol use and violence may primarily be over what is the prime influence on the violent behavior and whether alcohol is the necessary variable in an event culminating in violence. In a carefully designed research project involving all violent acts witnessed by or perpetrated on a random sample of citizens in a single town, Pernanen37 found that alcohol was involved in 54% of those acts but that the effect of other factors on the violent outcome varied even more than alcohol use by the victims and perpetrators. Some studies have demonstrated the missing effect by giving alcohol or substances purported to be alcohol to study subjects under single-blind conditions.28,38 The researchers then observed the effect of the subjects’ believing they had drunk alcohol as well as the context in which that alcohol had been drunk on the subjects’ behavior. There are limits on how far the cultural setting can influence behavior if alcohol continues to be consumed. Eventually, higher cognitive processes (e.g., attention, concentration) do become impaired. In the end, it is the combined effects of the pharmacological, situational, and cognitive factors that determine the outcome: feelings of being threatened, provoked, or insecure; intrapsychic components (e.g., depression, anxiety); and the influence of alcohol can all contribute to an act culminating in homicide. A similar critical appraisal could be done for the effects of psychoactive drugs on an individual’s potential for extreme violence. In one study, 81% of 275 homicide victims were tested for benzoylecgonine, the major metabolite of cocaine. Forty percent of the victims tested positive for this metabolite, a number much higher than that presumed in earlier studies.39 Although marijuana or opiates may inhibit violent behavior, phencyclidine, lysergic acid diethylamide (LSD), and amphetamines are seen as increasing the potential for violence. The changing patterns of drug us-

26


age and drug subcultures, especially in major American cities, need to be assessed at any given time. Drugs vary in popularity at different periods of time. For example, in the 1970s heroin injections were popular, followed by the emergence of cocaine. In the 1980s and 1990s, smoking crack cocaine emerged with its accompaniment of increased violence. In the 1990s, the increased use of methamphetamine began, and it continues into the twenty-first century. A related drug that has emerged is methylenedioxymethamphetamine (MDMA), which is known as “Ecstasy” and is a stimulant and psychedelic that induces emotional lability and surges of energy so a person can stay active for prolonged periods of time. Diverse drugs are still used and different patterns witnessed, such as smoking marijuana in a cigar (a blunt) along with consuming alcohol. All of these have different potentials for violence depending on the vulnerabilities of the individual, the social context, and whether the person is also involved in dealing drugs.

TEMPORAL AND ECOLOGICAL FACTORS Uncontrollable Environmental Factors In one case, a man called a rival, threatened to kill him, and announced that he was driving to this man’s home to carry out his threat. The police were called and were waiting for him, but the man never arrived. Meanwhile, a different police squad was called to help retrieve a car that had skidded off the highway into a snowbank. In the car was the man who had made the threats, with a loaded shotgun by his side; only because of environmental factors was he rendered unable to progress further on his mission. This rather mundane example of an impediment to a homicide illustrates the idiosyncratic variables affecting whether a homicide occurs.

Seasonal, Time of Day, and Other Periodic Factors Seasonal factors have been studied to determine the incidence of violence, including homicides, by calendar month. The UCRs indicate that most homicides occur in the summer, but the months with the highest homicide rates are July and December.13 These high-rate periods may simply reflect periods in which there are greater interpersonal contacts and social mingling (i.e., summer with its long days and good weather, December with its many holidays). Similarly, the day or time of day most likely for the occurrence of a homicide has been noted.40 Saturday evening and early Sunday morn-


27

ing appear to be the prime times, which would seem related to the most common time that people are exposed to each other. The search for periodicity in homicide has led to recurring biological factors in homicides. One phenomenon referred to is “premenstrual frenzy,” which is discussed in Chapter 2 (“Biological Factors in Homicide”), under the heading “Premenstrual Syndrome.” Humans’ fascination with the moon has never diminished, even though astronauts have now walked on the moon. In mythology, stories, and song, profound powers have been attributed to the moon. “It is the very error of the moon; she comes more near the earth than she was wont, and makes men mad,” as Shakespeare expressed it in Othello.41 The reasoning is that the human body responds to the moon much in the same way as the sea, especially considering that the body is made up of 80% water, like the earth. The influence of the moon would also be induced via changes in the geophysical environment in weather, earthquakes, and electromagnetic fields. One investigation from this perspective in Dade County, FL, showed a correlation between the lunar phase cycle and homicides and revealed that homicides peaked during a full moon.42

CULTURAL VARIABLES In addition to discussions of different homicide rates within different social groups in a particular society, it must be recognized that cultural settings also influence homicide rates. This is true even when making comparisons among industrialized countries, let alone between industrialized and less industrialized countries. Within the industrialized nations, the United States has the highest murder rate; Japan and Austria have the lowest rates. Some economically developing nations do have higher homicide rates than the United States; the homicide rates per 100,000 in these countries are as follows: Colombia (56.3), El Salvador (38.8), Jamaica (37.2), and Mongolia (24.6); however, it is not clear whether these rates exclude war-related activities.43 The homicide victim rate for most industrialized countries is usually below 3 per 100,000. However, from 1980 onward, the homicide rate in the United States has been more than twice that figure, falling somewhere between 5 and 10 per 100,000. When the rates are broken down in terms of age groupings, the disparity in the rate in the United States compared with most European countries becomes even more glaring, because the rate for young males in Europe lies between 1 and 2 per 100,000. As far as can be determined from existing documentation, the

28


murder rate in the United States in the twentieth century has never been as low as 3 per 100,000. It is informative to compare the numbers of homicides in different countries. Interpol data from 1999 to 2000 show rates of homicide per 100,000 for Europe to be 1.59.44 Certain countries stand out, such as Lithuania (10.62), Estonia (10.61), and Latvia (6.47). Russia is listed at a rate of 22.05 and South Africa at 55.86. At the lower end are Spain and Switzerland (1.12), Sweden (1.11), Japan (1.05), and Norway (0.05). Explanations of this striking disparity vary widely. One is that Americans are more tolerant of aggression and violence, including violence with firearms, which are considered to be ever present in American society. (See earlier discussion of gun-related homicides in the section “Method of Killing.”) Another theory is that American society does little to discourage physical aggression among its young male population. Still other theories stress the effects of the media and network television violence on America’s tendency to be a violent nation. The conclusion of some researchers is that the United States is a society that promotes and accepts the extremes of personal aggressive violence but is hypocritical about other standards, such as sexual deviance, as witnessed in discriminatory treatment of those with diverse sexual practices.45

SOCIOLOGICAL VARIABLES Changes in Lifestyle as Increasing Risk for Victimization Some researchers see the question of who is likely to meet death by homicide as a subquestion in the context of a broader, sociological one: why is there more crime and violence in modern society, when material conditions for living are comparatively better than in preceding historical periods (an interpretation of history with which not all agree)? Those who view these questions from a lifestyle approach see the quality of people’s lives and interactions with others in modern society as not being improved and consider the acquisition of more material goods as irrelevant. In fact, some argue that a milieu for increased violent behavior has been created, referring to variables such as the following, among many others: • The breakdown in family structure • Increased number of families in which both partners work outside the home • Minimal time at home • The increasing number of meals people eat out


29

• Increased numbers of people living in a state of isolation • Social lives increasingly centered outside the home • The pursuit of entertainment and a social life based on endless shopping malls that are open 24 hours a day, 7 days a week • Increased exposure to violence in various forms of media As a result of these factors, people are placed in locations where they can more often become targets of violence. More perpetrators are then stimulated to capitalize on the more readily available targets.46 Implicit in this theory is that anything that increases the contact between human beings will increase the risk of a homicide. If, because of contemporary lifestyles, people are on the move more often, they are consequently more exposed to the possibility of violence. Hindelang and colleagues47 posited a theory on what factor increases the risk of being victimized because of more contact with people under certain conditions. As a common example, an increased risk of violence accompanies roles certain people play, such as a single person being more likely to go to diverse places and be exposed to unknown people more frequently. It is not only that single individuals do more things alone, and that they do them alone more frequently than before, but also that they expose themselves more to the different phenomena and contexts that this lifestyle provides (e.g., bars, clubs), which in turn increases their risk of becoming a victim of violence. Another factor increasing the risk of homicide is one’s place in the social structure. For example, a cabdriver going on “runs” into a druginfested part of a city increases his or her risk of being killed, just as does a college student’s dabbling in selling drugs. Constraints of race or social class operate within this framework. Of course, a person may try to set limits on these activities by avoiding certain exposures, but depending on the factors just discussed, certain risks may be unavoidable depending on one’s circumstances. Some research has specifically tested these hypotheses. One group evolved an index called the household activity ratio. Married households where the female spouse worked and households headed by a single adult with children were studied in proportion to the total number of households of all kinds. The former group was seen as being in a higher-risk category for being a homicide victim; in fact, the ratio provided by comparing the first two groups with the total households in the entire United States for 1947–1974 had a positive correlate with the rate of homicide.48 A variation on this theory is to see the ratio between the first two groups and the total households as signifying economic inequality, which in itself can lead to an increase in homicides.49 For households in which

30


both partners work, that household has double exposure to risk in that both partners leave the home each day; for households headed by a single adult, members of that household are more likely to have a lowered socioeconomic status and be exposed to poverty-like conditions and the accompanying offshoots, with such economic inequalities then placing that household at a greater risk for homicidal violence. These adverse economic conditions have a general adverse influence on individuals and families in a spiraling effect. A common example is the number of people in the lower socioeconomic group living in the inner parts of large cities with limited work opportunities, an increasing minority population concentrated in public housing, and illicit drug dealing that is perceived as a way out with its increased risk of violence.50 In an analysis of individual homicides in New York City, the impact routine activities have on the rate of stranger-committed homicides was confirmed.51 According to this theory, if males are away from home more than females as a part of their daily routines, then they are more likely to become victims of homicides, a tendency that was confirmed in the New York analysis. Continuing with this hypothesis, white individuals would then be more likely than other races to become victims of stranger homicide because their jobs are more likely to take them farther from home or require travel. With increasing numbers of women away from home at present, they are more likely to become victims of stranger-committed homicides.

SUBCULTURE OF VIOLENCE The essence of the theory of a subculture of violence is that certain subgroups in American society hold values conducive to violence and that the socialization process in those groups not only accepts but promotes such behavior. The initial formulation of this theory was not restricted to homicide but rather was applied overall to criminal behavior in which such behaviors were seen as being learned. Through interactions with individuals who approve of crime, norms develop that are conducive to crime within certain subgroups. For example, a subgroup may be more tolerant of violence and likely see it as a legitimate means for the expression of aggression or violent tendencies. If a homicide results, it is seen as unfortunate for both parties but simply an inherent risk in the way life is lived in the subculture. Such an outcome is not seen simply as two individuals interacting but as a reflection of some social groups more often resorting to violence that in turn increases the likelihood of homicide.


31

The exposure of members of these subgroups to violence can be intense, from the enormous violence witnessed on television and in films to the subculture’s connection of male identity with expressions of violence. Weapons are readily available for those who believe they need to “even things up” to express their aggression. Going along with the acceptance and promotion of these values is the hypocritical policing of sex in magazines, music, and television in society, which gives a strange mixture of messages in which sex is seemingly regulated but violence promoted. One hope for members of this subculture is that the violence will be channeled into other options (e.g., competitive athletics among these groups). Another hope is that those who are observers and vicarious participants in the subgroups can mitigate the degree of their own need for direct aggression. Of course, the linchpin of the theory that a subculture of violence leads to increased homicides is that those subgroups who have high rates of homicide actually do possess a different set of norms with respect to violence. The reverse—that high homicide rates validate such a culture—cannot be argued; rather, independent empirical validation of the theory is needed. Measuring attitudes toward violence is difficult; even then, expressed behavior may not match verbalized values. For example, almost no one directly states that he or she approves of violence, let alone the killing of someone (with the exception of those involved in political killing and ethnic cleansing). Rather, it is the covert way in which different attitudes toward violence are expressed that would distinguish one group from another. As a result, it is not easy for groups or individuals to be classified as part of a subcultural group or not. It is more likely that they could be classified along a continuum. One group may approve of a good deal of criminal behavior and stand in opposition to many societal values, yet only approve of certain selected killings. In between may be groups that have a mixture of values, some of which may be conducive to violence and others not. On the other end of the continuum may be a group that has values approving of more minor antisocial behavior, such as white-collar or bureaucratic violations, but that sees a homicide that results from such activities as outside the accepted norms. Such continua exist for groups as well as individuals within the group.

Strain Theory Contemporary proponents of the subcultural theory have made many different arguments from those mentioned previously. These pro-

32


ponents rely on a strain theory, in which it is hypothesized that when individuals cannot achieve their goals (e.g., monetary gains) through legitimate channels, they are more likely to resort to crime. When involved in frustrating situations, people are at increased risk to act out their anger and rage. Strain theory overlaps with a mixture of interpersonal stress and intrapsychic conflict that psychiatrists deal with. The original writings within this theoretical framework stemmed from the well-known sociologist Robert Merton.52 Subsequently, his work went in various directions, such as focusing on the different goals people pursue, the barriers to achieving such goals, and how and why some individuals operating under strain resort to crime. Some strains are more conducive to homicidal levels of violence. In the context of interactions with others, an individual with poor controls and low social skills is likely to act out in a violent manner. A feeling of being oppressed or treated unjustly would leave an individual more vulnerable to react to these strains with protesting or rebellious behavior. A divisive impact in the work or school milieu would heighten these feelings and leave a residue of irritability and proneness to take offense. Clinical evidence shows an increased prevalence of irritability in major depressive disorders associated with a greater likelihood of suicide.53 The presence of irritability and difficulty modulating anger in borderline personalities is relevant as well. These findings could be raised for homicides. Subsequently, interactions with other “strained” individuals may provide a social support setting and increase the likelihood of group behavior becoming criminal behavior in which violence comes to play a part.54 Parental absences or inadequacies can further accentuate strains in a juvenile or young adult population. A recent addition to the criminal law has been the category of hate or bias crimes, in which the perpetrator’s state of mind and motivation toward the victim are the key. The very same act, such as a homicide, is distinguished from a “parallel” crime simply by virtue of the requisite mental state accompanying the act. Thus, a bias homicide occurs if the person killed was selected based on discrimination or racial animus. In 1990, the Federal Hate Crime Statistics Act provided that the FBI would track offenses in which violent attacks, arson, intimidation, and property attacks were based on “manifest evidence” of prejudice based on race, religion, sexual orientation, or ethnicity. The purpose of the act was to aid in identifying criminal acts in which bias was present. By 1994, a crime bill had been directed to the U.S. Sentencing Commission to promulgate guidelines to enhance penalties for such offenses. Many states have adopted similar legislation to enhance penalties for acts that would have received a lesser penalty otherwise. However, determina-


33

tion of the motive in such cases may be difficult and involve a search involving complex interpersonal dynamics. The result of opening this area at sentencing may be confusing. Current efforts to conceptualize and measure hate may prove helpful.55

Other Theories Others have argued that the strain theory is not sufficient to explain certain subcultures of violence. A revision has been offered that some individuals in groups selectively endorse, as well as distort, values of the dominant culture. Matza56 stressed that in some subcultures, the values of toughness, excitement, and pleasure seeking—values taken from the culture at large—are emphasized. Within the context of these prioritized values, violence and homicide are more likely to emerge from the increased risk taking and impulsivity that may be present. Because members of these subcultures see these values exploited in society at large, such as in criminal law procedures or in the corruption of how some businesses operate and rationalize their behavior, they feel they are justified in adhering to their own values, even if it leads to homicide and the exclusion of other values promulgated by society.

FELONY MURDERS In some epidemiological studies, the concept of primary homicide (again, meaning a death resulting from a more or less spontaneous confrontation or fight between individuals or even groups) has been distinguished from the category of felony homicides, meaning killings taking place in the context of another felony (e.g., robberies, sexual killings, executions, violence between rival gangs, murders planned to achieve a certain goal [one gang sending a “message” to a rival gang]). The English common law was originally described as holding felons strictly liable for all deaths caused in the course of felonies. This rule became the rule by default in American jurisdictions and has been described as a myth.57 However, various state statutes were enacted in the nineteenth century that began to incorporate these ideas, although with different principles of liability for a felony murder. Although the number of felony homicides is significantly less than the number of primary homicides, with some estimating their prevalence at only about 20%, they continue to receive much more attention in the media than primary homicides because of their dramatic aspects and because they have greater appeal in terms of being fictionalized. When homicide rates be-

34


gan to increase in the 1960s, the proportion of felony homicides also increased. The result was a drop in the clearance rate for homicide because of the increase in these types of felony homicides, which do not have as high an arrest rate. Those who have studied felony homicide often see it as an impersonal and predatory type of murder in contrast to murders committed in the heat of passion by intimates. In a classic work discussing felony homicides, Dietz58 listed their distinguishing characteristics: 1. They are usually enacted by experienced, habitual, or career criminals. 2. The planning and premeditation of a felony homicide are critical parts of the act. 3. Felony homicides may involve multiple victims and multiple offenders. 4. Victims may be forcibly or physically restrained. 5. A specific antecedent (e.g., a relevant argument), as may be present in homicides between intimates, is often lacking. 6. The role of the perpetrator of a felony homicide is hypothesized, tested, and incorporated or rejected as part of the self-concept of the perpetrator. Robbery homicides are the most frequent type of felony homicide; a bank robbery is only one example of this. Dietz hypothesized that robberies occur in distinct stages. First is the planning stage, which usually occurs within a few days of the actual robbery. Sometimes those planning the robbery may specify killings during the robbery, but in other cases not all those to be involved in perpetrating the robbery are informed of this aspect of the plan, whether it is conceived that a killing will occur as a direct consequence of the plan or only as the need arises. One pattern may be that two or more of the group members are informed about the intended killing or make the decision they will do so if needed, with the other members acquiescing in such a decision. The victim or victims are usually regarded as someone impersonal or someone undeserving of respect. During the posthomicidal period, the members may talk about a killing that occurred in the course of a robbery and, in some cases, joke about it. Other variations on felony homicides occur within the context of planned executions, contract killings, or revenge killings and are related to street gangs or planned killings between rivals in organized crime. In some of these situations, a professional, referred to as a “hit man,” is hired to do the killing.


35

Although technically considered felony murders, homicides committed during a sexual assault (see Chapter 10, “Sexual Homicide”) differ in many aspects from these types of felony homicides. It should be noted that the concept of felony homicide has also acquired a series of legal meanings connected with cases of who might be tried for a homicide. As this concept was initially conceived, the accomplice of the perpetrator of a homicide that occurred during a bank robbery would also be charged with homicide. That concept has now been expanded so that the person driving the getaway car from a robbery in which a homicide was committed, even though he or she was not present at the killing, would be legally charged with felony homicide. As a further expansion of this concept, if a bank security guard kills one of the bank robbers in the course of the felony or attempted felony, the other members of the group attempting to carry out the felony could be legally charged with being responsible for that death. Apart from sexual killings, psychiatric issues are much less frequently raised with this group of felony murderers than in the primary type of homicide among intimates. Those involved in organized gangs—whether juvenile gangs or more advanced criminal organizations—or those controlling the gang do not wish to explore the intricacies of their psychiatric state at the time of the offense. Such inquiry is not part of their group ethos. Second, their own self-concept is such that they do not wish to be thought of as being deviant in a psychiatric sense, which would be unacceptable, even though social deviance is acceptable. Hence, many of the complex dynamics involved in their lifestyles are never explored.

MULTIPLE-VICTIM MURDERS Mass and serial murders have gained increasing publicity in the United States. Whether they have in fact been more prevalent in recent years, or whether American society is now simply becoming more aware of these types of killings through improved detection and investigation, is not clear. The Bureau of Justice classifies three types of multiple-victim murderers: serial murderers, mass murderers, and spree murderers.

Serial Killers A mass killing is defined by the FBI’s Behavioral Sciences Unit as the murder of four or more victims in a single episode, although some researchers prefer a threshold of three. A serial killer is distinguished from a spree killer simply by the time elapsed, such as whether there has been

36


a “cooling off” period between attacks. Serial killings can extend over months or years between the killings, whereas spree killings are seen as a binge extended over a period of hours or days in a continuous pattern. The names of several notorious serial killers come to mind under this definition, each exemplifying different types of killings:59 • Jack the Ripper, who committed several sexual murders in nineteenth-century England. • Ted Bundy, suspected of killing as many as 36 young women across the United States. His saga ended with his conviction in Florida for murdering two female Florida State University students and his subsequent execution. • Jeffrey Dahmer, who killed and cannibalized adolescent males, was convicted and later murdered while in prison. • Two men who ran a “little house of horrors” in Philadelphia, PA, where women were held in the house, abused, tortured, and eventually murdered and then cannibalized.60 • Various individuals who killed elderly people in rooming houses or boarding home situations and then appropriated their pension or Social Security checks. • Dorothea Montalvo Puente, in California, who killed nine elderly people who were her tenants for their Social Security checks and buried them in the lush yard around her house.61 • Aileen Wuornos killed seven men in a 13-month period after having sex for money with them. She pleaded self-defense on the basis of rape. The movie Monster was about her. • Albert DeSalvo, commonly referred to as the Boston Strangler. • David Berkowitz, who had the nickname “Son of Sam” and killed seven women. • The “Hillside Stranglers” of California, who were the cousins Kenneth Bianchi and Angelo Buono. • Richard Ramirez, the “Night Stalker” of Los Angeles, CA, who continued to be sought out by young women even after his imprisonment, an interesting theme in its own right. • Wayne Williams, a black male in Atlanta, GA, who carried out repeated killings of adolescent black males. • Henry Lee Lucas, who confessed to killing 350 people between 1975 and 1983. He subsequently recanted these confessions, which raised challenging issues about the veracity of these individuals and the difficulty in getting scientific data for profiling them. • John Wayne Gacy, who sexually molested his male victims and was convicted of killing 33 people.


37

• Ted Kaczynski, an intellectually gifted ex-professor who sent explosive devices in the mail in protest against a technological society. • Other high-publicity cases that have been written about include those of Juan Corona, Herman Mudgett, and Wayne Henley.

Incidence of Serial Killings It is almost impossible to obtain an accurate figure on the number of people who have become victims of multiple killers. Estimates range widely from 3,000 and up and for serial killers from less than 100 to over 500. Part of the difficulty is definitional. Some would exclude mass killings that are done for financial gain; others would exclude those within a family context (familicides). Whether killings involving mobs or drug dealings that involve multiple victims should be included is another issue. Separate records are not kept by the FBI or the Centers for Disease Control and Prevention on serial killings. The only semblance of record keeping is by the FBI’s Behavioral Analysis Unit (formerly the Behavioral Sciences and Investigative Support Unit) at the National Center for the Analysis of Violent Crime, which relies on newspaper wire services.62 Valid estimates are also contaminated by the mass media’s tendency to overplay such a news-catching phenomenon, which is then followed by a sequence of panic reactions on the part of legislators and other policy makers.

Patterns Seen in Serial Killings One difficulty in understanding serial murderers’ patterns is that these killers often drift around the country killing people, making it very difficult for them to be identified because their patterns are not picked up in one community. In addition, some serial killers select victims who are themselves drifters and not stable members of a community, and in that sense it is difficult to sort out the killer’s pattern. The victims in these cases are often people on the margins of society themselves, such as prostitutes, people with alcohol or drug abuse difficulties who are living in public settings, or runaway children and adolescents. Because the victims may be strangers, the usual sources of information to detect the killer, such as members of the family or friends, are not available or produce almost minimal leads because the victims may have been gone from home for some time. Other characteristics have been listed in an attempt to delineate more clearly the patterns of serial killers. The killings often occur at different times and seem to have no apparent connection to one another.

38


Some serial killers wander about the country and pursue victims, whereas some are “geographically stable” and stay in their home locale, such as Wayne Williams in Atlanta, GA, or John Wayne Gacy in Chicago, IL, and kill there. Some serial killers may wander simply to make it more difficult to catch them. Determining which subgroup of serial killers, if any, meet clinical criteria for different types of psychotic disorders is also needed. The understanding is that some of these individuals are in the midst of a full-blown psychotic disturbance (e.g., some type of schizophrenic disorder, a major affective disorder), whereas others show evidence of personality disturbances. Serial killers display various behaviors in connection with their killings in terms of how they treat their victims (e.g., using blindfolds, attacking their faces, tying up the victim), dispose of their bodies (e.g., dismembering the body, placing the body in specific positions), and “memorialize” the act (e.g., some take souvenirs from the victim or scene of the crime). Holmes and Holmes63 obtained some information on these variables from interviews with serial murderers and saw their principal motivating factor as a pleasure in killing or the gratification in holding the fate of others in their hands. Blindfolds appear to have significance in terms of whether the perpetrator is comfortable enough to have the gaze of the victim fixed on him or her during the various types of assaultive behaviors that culminate in the killing; the perpetrator could use the blindfold on the victim or on himself or herself. It is hypothesized that attacks on the victim’s face, particularly to blot out the eyes, may be a similar way of eliminating and controlling the shame that partially breaks through the perpetrator’s defenses against affect. Similarly, it has been thought that oral sex that is accompanied by blindfolding in connection with these killings suggests a stranger-perpetrated offense, which relieves the perpetrator from having to view the victim as a fellow human being. The blindfold would serve little purpose with someone who is already known. Use of weapons as part of the torture pattern may involve ritualistic aspects. If weapons are used, a serial killer usually selects one that requires personal contact between the killer and the victim because of a desire to touch or terrorize the victim or because the act may be degrading in terms of the method used, such as strangulation, blunt instruments, and so on. Finally, if a dismemberment takes place, it is seen along the continuum of adding power and control over a powerless victim who is simply carved into pieces or repeatedly stabbed. Similarly, bondage occurs in terms of keeping victims helpless or in a degrading position where they can be tortured or mutilated before their remains are disposed of. Torture and killing of political prisoners or those held


39

in captivity raise different questions that are unresolved. Differences in the individual personalities and group influences that operate in such situations differentiate who participates in such acts and who does not. Again, different possibilities exist with respect to the manner in which the bodies of victims are disposed of, reflecting different types of serial killers. It is thought that serial killers who dispose of the bodies of their victims so they are not likely to be discovered would more likely be in the category of a lust, thrill, or power-control type of killer. None of the serial killers appear to be concerned about the families of the victims in terms of how relatives must live with unresolved anxiety and anguish surrounding the disappearance of the bodies. Because victims are seen as nonentities by the perpetrators, it does not cognitively register for them to put themselves in the position of the families of the victims. For some types of serial killers, great care is taken in disposal of the body because that is when they could be most vulnerable to discovery. Thus some victims are buried on the perpetrators’ property to minimize the risk of discovery. Another phenomenon seen in some serial murderers after their apprehension is that they confess to far more murders than have actually been committed. The police may thus “solve” many unsolved cases in this manner. The dynamics of the individual who confesses to more serial killings than have actually occurred are interesting in that they touch on the murderer’s narcissism and grandiosity.

Characteristics of Serial Killers It is difficult with the limited knowledge available to capture exactly what the characteristics of a serial killer are. One problem is that descriptions of serial killings vary among experts. Many workers in the field describe a serial killing as when one or more individuals commit a second murder or a subsequent number in which the victim is “relationshipless” to the killer.64 Others take issue with this requirement, allowing that a relationship might exist between the serial killer and some of the victims. Examples given are those of a professional person in a relationship with a client or patient who subsequently kills him or her, or a landlord who has a series of tenants that are killed over a period of time. (The possibility of people in more intimate familial roles carrying out killings is discussed later [see “Familicide” section]; familicide involves a different conceptual basis than that in impersonal killings.) Most investigators feel that serial killings are not done for material gain but perhaps are associated with a desire to have power over the victims. Some killings may be carried out primarily for the sadistic grat-

40


ification from the act of killing itself. A subgroup are seen as a response to a pathological need for attention and sympathy. Other homicides seem to be connected to quasi-cult activities that are ritualistic and brutal, such as those perpetrated by Charles Manson and his gang of followers.65 Other descriptions, instead of focusing on the killer’s motive, stress that the killings may have symbolic value when carried out against victims who are perceived as being defenseless. Another theory is that the killings are the perpetrator’s way of alerting others to the victims’ plight; in essence, the victims are perceived as powerless and without status within their immediate surroundings (e.g., vagrants, migrant workers, children on the run, homosexual individuals, prostitutes, elderly women, captives). More specific documentation is needed to determine whether there is, in fact, a selectivity in terms of victim selection going on with different serial killers. Similarly, some serial killers see themselves as having a mission to rid the world of certain types of individuals. Although this borders close to the class of political killings, in which the killers are similarly motivated to eliminate a certain group of people, serial killers’ victims are chosen simply on the basis of their own cognitive convictions that it is their duty to rid the world of undesirables or a particular class. This motive would appear to be behind some of the killings in which prostitutes or homosexuals are selected as victims; friends of the perpetrator are often stunned to learn the identity of the perpetrator because he or she may come from an “upstanding background.” These types of killings also present some overlap with the new category of hate crimes. The group of “sadistic serial killers” who perpetrate erotized murders have their own characteristics, in which mutilation, torture, dismemberment, or other types of traumatic activity are often connected with the killing. Such killings are not all sexual murders in an overt sense, for a type of hedonistic pleasure can occur with some who perpetrate this type of killing. Some also become labeled cult killings. Neurotic conflict possibly underlies the actions in these cases. Whether having total power over a helpless victim always conveys some type of neurotic basis is not at present a position that can be confirmed or disconfirmed. A related question is whether the serial killer’s behavior tends to escalate (e.g., decreasing time between killings, escalating sadistic brutality in connection with the killings) or whether this escalation only occurs in a subgroup in contrast to serial killers whose behavior does not escalate. The escalation may reflect an increasing momentum and poorer controls, and it also increases the possibility for detection. In Godwin’s66 study of 107 serial killers and 728 victims, it was noted that


41

despite differences in social class, IQ, or education, those who remained undetected for prolonged periods of time were usually cunning and skillful at presenting themselves. Such cleverness placed them beyond suspicion and made them difficult to apprehend. In that sense, they were methodical and organized as they went about their business of killing.

Role of Fantasies The role of fantasized images and objects connected with victim selectivity has specific psychiatric significance. There has long been a belief that the serial killer seeks out some type of idealized victim to pursue. Although this may be so in their fantasies, the records from serial killers indicate that what goes on is a type of depersonalization process regarding the victims selected. In their compulsion to carry out a homicidal act, a more generalized object will often do. In fact, the nature of the process is such that the victim selected is not only depersonalized but viewed with some type of contempt or repugnance that seems to facilitate the violence being carried out. This degrading process of the victim was described by one serial killer as follows: “So, he mentally transforms them into hateful creatures, because, in the twisted morality of his own making, it is only against such that he can justifiably and joyfully inflict his manifestly hateful deeds of violence.”63 Determining the origins and vicissitudes of the fantasies to carry out acts of violence in the process of gaining control over victims is a worthy clinical topic in its own right. Some of these fantasies emerge in a nascent form in which they do not originally include the imagery of carrying out the homicidal act but evolve over time. Perhaps the fantasies are elaborations on certain actual experiences or are stimulated by “experiments” conceived in various types of movies or magazines. Some of this material may be integrated into actual sexual experiences or into masturbatory fantasies. A crucial shift occurs when the potential perpetrator begins to elaborate on these fantasies using actual people encountered, perhaps even rarely, in his or her daily life. Why, at a certain juncture, a small group of individuals begins to become discontented, if not anxious, about the frustrations of simply confining themselves to a fantasy level is a key question. At that point, the potential perpetrator plays out in his or her mind how the fantasized activity might actually be committed. This involves the distinct possibility that his or her control systems have become less effective. Once the potential perpetrator actually considers carrying out some type of violence against another human being, a crucial step has been

42


taken. At first victims become depersonalized and then are viewed with contempt or anger. This transition seems to be connected with a process in which the potential victims are seen as people who are frustrating the potential perpetrator by not being available for the perpetrator’s purposes when wanted. This defensive elaboration is crucial to the perpetrator’s being able to carry out the type of violence anticipated, for if he or she recognized the victim as simply being an innocent person who has actually done no harm or wrong whatsoever to the perpetrator, it would be difficult for the perpetrator to carry through with the act. There are also indications that when it comes to the final enactment of killings in this sense, the act is related to a sense of utter hopelessness and despair. To a clinician, this may signify the possibility that a severe degree of depression is related to the act of killing. Thus it would seem that the perpetrator puts fantasies into action at the point at which there is an onset of a major depression, if not the threat of suicide. It is in response to the anger and rage they feel that perpetrators contemplate the final “solution,” with the idea that things will be made better by acting out violent tendencies toward helpless victims—the actor survives at the expense of a person who is seen as being worthless and degraded. In that sense, the perpetrator’s self-esteem is temporarily reestablished without his or her experiencing any personal feelings. In light of these proposed dynamics, there is also a possible explanation for the type of sadistic and degrading acts perpetrated in some cases to prolong the victims’ suffering; such degradation allows perpetrators to view themselves as all-powerful and the others as existing solely for the purpose of allowing someone who is superior to use them in such a manner. Hence, all manner of subjugating and torturing a victim may be seen as acceptable. Once there is relief from a state of utter despair and isolation by way of torturing or sadistically using a victim, the victim can be killed and disposed of like any other type of disposable object, because the victim lacks any personal meaning.

Mass Killings A mass murderer as noted is one who kills three or four or more victims at one time as part of an ongoing killing event. It is distinguished from serial killings, which are seen as having a 30-day interval between killings. Many mass murderers may die at the scene as a predetermined part of the act or in a shoot-out with police. Infamous names occur in this regard, each of which would warrant an investigative study in his or her own right. A few examples follow (apart from the political terroristic killings such as the attacks of September 11, 2001).


43

• The worst mass killing in earlier times in the United States occurred in 1955 when an explosion on a passenger plane killed 44 people in California. It was later revealed that the son of one of the passengers on board the plane had planted a bomb to collect life insurance on his mother. • In 1966, Richard F. Speck sequentially murdered eight student nurses in their Chicago, IL, apartment. A ninth hid under a bed and was able to report the incident. • Also in 1966, Charles Whitman climbed a tower on the campus of the University of Texas and began shooting the people below. He killed 16 people and wounded 30. He was a former Eagle Scout, Marine, and engineering honor student. The day before he had killed his mother and wife. • In 1983, 13 businessmen and gambling dealers were shot in a Seattle, WA, gambling club. • In 1984, James Huberty entered a McDonald’s restaurant in San Ysidro, CA, shooting everyone in sight, with the result that 21 people were killed and 19 wounded. He was eventually killed by police sharpshooters. • Patrick Henry Sherrill, a mail carrier who was having trouble with his supervisors, opened fire in a crowded post office in Oklahoma in August 1986, killing 14 fellow workers and injuring 7 others before committing suicide. • In 1991, another disgruntled postal worker killed three coworkers and wounded six more in Michigan. • Another major mass murder in the United States occurred on October 19, 1991, when a single individual, George Hennard, entered a cafeteria in Killeen, TX, with two semiautomatic pistols and killed 22 customers and wounded 23 more. It is ironic that the next day, the U.S. House of Representatives voted 247 to 177 to defeat a bill that would have banned 13 types of assault weapons and the high-capacity ammunition clips that make these guns so lethal.67 • On April 19, 1995, the bombing of a federal building in Oklahoma City, OK, by Timothy McVeigh killed 167 people, for which he was found guilty and executed. A question is whether McVeigh’s type of political killing, along with the 3,000-some deaths on September 11, 2001, should be considered a mass killing—which it literally was. However, since the motives and classifications of terrorist killers seem to differ significantly from the above group, they are usually discussed separately.

44


Familicide A variation on mass murder is the killing of an entire family by one member. I call this familicide. This type of mass murder has a different type of psychodynamic and psychiatric significance, because it involves an individual killing his or her own family members or loved ones en masse versus killing anonymous people. This distinction is not often made, which blurs an attempt to understand the psychiatric differences in such overt acts.68

Medical Mass Murder Using prisoners for purposes of experimentation, based on certain political or racial beliefs, is referred to as a medical mass murder,69 such as was seen under the Nazis. A variation on this is the cases reported from medical institutions or nursing homes in which an individual supposedly carries out acts of kindness or mercy killing by “putting away” the elderly or those with some disability. These are not acts done in terms of revenge or overt anger but rather are seen by the perpetrators as altruistic acts. Sometimes the acts are carried out on those in the throes of a terminal disease, through overdoses of lethal drugs. This topic is discussed further in Chapter 7 (“Masochism and Homicide”).

Spree Killers In spree murders, a killer goes on a rampage. It is usually defined as three or more people being killed in a 30-day period. The events may all be a part of a contiguous series of killings in which the rampage simply incorporates the killing as part of it. An example would be a series of armed robberies in which the perpetrators shoot and kill various people as they drive away. Other killings may take place in this context in terms of stopping at various places along the way subsequent to some type of felony and killing at those locations as well, such as at gas stations, restaurants, and so on. There is usually a path of violence occurring over several days. For some the killing is part of a plan to go on such a rampage and to evade being caught. The homicides committed by Bonnie Parker and Clyde Barrow (“Bonnie and Clyde”) and by Charles Starkweather are some examples, as are the more recent sniper shootings by 18-year-old Lee Malvo and 42-year-old John Muhammad in the Virginia, Maryland, and Washington, DC, areas in 2002 that killed 10 people and wounded three.


45

Terroristic Mass Killings Terrorist killings are another form of mass killings. Such killings come in many packages and often reflect a political, religious, or social reform agenda. These killings do not follow the standard rules of countries at war, because the victims are usually innocent and only symbolic of what is seen as needing to be destroyed. If the terrorists themselves are killed, or suicide in the process, they view their deaths as necessary, if not honorable, in the service of a just cause. Although one side may view the acts as heroic, the other views them as despicable and acts of outright murder. There are innumerable examples of terrorist killings throughout history—more recently, the September 11, 2001, attacks in the United States. Some terrorist killings occur inside the domestic side of a country’s politics, whereas others are international in scope. The majority of studies of such acts have come from the perspective of political scientists, sociologists, or social psychologists, with some philosophers viewing the acts as manifestations of evil or depravity. In many cases, a disparate cultural and religious background needs to be considered. A deficiency of in-depth psychiatric knowledge exists about most terrorists who engage in such killings. The situation is one in which attempts are made to reconstruct the mental state of the perpetrators from disparate sources because there is rarely direct psychiatric material available. Jerrold Post70 is one of the few psychiatrists who have pursued the understanding of terrorists, which involves “assessing leaders at a distance.” The goal is to provide a type of conceptual framework to identify the adaptive style, cognitive functioning, attitudes, affect, and interpersonal predispositions of terrorists. Such profiles are then by necessity based on information and sources outside direct participatory psychiatric assessments. Because terrorism is rarely the act of a sole person, there must be some understanding of how membership in a particular group has evolved. Group membership taps diverse emotions of loyalty, dedication, rage, revenge, guilt, and the possibility of betrayal. In the 1990s, various incidents of domestic terrorism occurred in the United States before the September 11 attacks. One prominent example was the siege in Montana between the FBI and the antigovernment Freeman group. Another was the confrontation between the government and the Branch Davidian sect led by David Koresh in Texas.71 Koresh had changed his name from Vernon Wayne Howell to Koresh, which was the Hebrew name for Cyrus, the Persian king who allowed the Jews to return to Palestine. Through various machinations, Koresh took command of the Waco section of the Davidians and armed the

46


compound in preparation for the apocalypse, the final battle between the forces of good against evil. In that battle, the world would be consumed in flames, and Koresh would return as Christ; those who believed in him would find paradise in the New Jerusalem. At the compound, Koresh lived among a harem of young girls who believed he was Jesus Christ and whose duty was to serve him sexually. Married men offered up their wives and daughters to him. The Bureau of Alcohol, Tobacco and Firearms and the FBI misread the situation as one of Koresh holding hostages rather than seeing them as committed believers. The result was a 51-day siege. The government viewed Koresh as a psychopath, whereas he seemed closer diagnostically to a borderline individual who was paranoid and delusional at times. The final attack by armored vehicles and tear gas led to a fire, with Koresh and 75 followers dying. The subsequent internationalist terrorist act of seizing domestic airplanes in flight to crash them into the World Trade Center and the Pentagon on September 11, 2001, by Muslim terrorists from al Qaeda eclipsed these earlier domestic events. Most studies of terrorist acts focus on such topics as why the acts occurred or how a particular organization was able to carry out the acts. Such events are followed by discussions and committees to investigate these matters and to assess the consequences of the acts on a society or political system. Unfortunately, any in-depth psychological analysis of the perpetrators is usually omitted. This omission may result from the minimal knowledge available about the perpetrators or from a lack of interest in that aspect of the killings in contrast to the big picture. Rosenfeld72 noted how qualitatively different such acts of terrorist killings are from other kinds of homicidal violence that criminologists study. Terrorist killings do not jibe easily with customary theories of violence such as social control theory or a life-course approach. Hypotheses about terrorism and terrorists are often not testable. Rather, they reflect partial analyses of extremist actions from the political left or right. The difficulty in explaining terrorist acts that result in multiple deaths has been put in terms of explanations lacking a logical comparability and a specification of how the variables relate to each other. There is an absence of knowing how to rank the variables in terms of their explanatory power.73 An explanatory model would have to address what encourages or inhibits individuals in participating in such actions, because the great majority of people in a given social milieu do not join such organizations or engage in terrorist acts and have no desire to do so. The similarity of the mind-set of terrorists to that of certain extremist religious sects has often been observed and has been referred to as killing


47

in the name of God, Jehovah, or Allah. Part of this occurs through a progressive isolation from others, accompanied by a professed loyalty and a commitment to a cause that may involve people getting killed. The missionary aspect to convert others to the cause reinforces the terrorists’ own beliefs. According to Stern,74 only two characteristics are crucial for distinguishing terrorist violence from other kinds of violence: the acts are aimed at noncombatants, and there is a dramatic quality to instill fear or dread. Often a sensed need for vengeance is used as the justification for innocent people being killed. In turn, the belief is that the perpetrators are acting virtuously in a just cause and that their own deaths will be avenged by their compatriots. This contributes to less hesitancy in participating in acts of mass killing. When some of the group die in the terroristic acts, the remaining members of the group draw closer, which may allow taking even greater risks. Survivor guilt may make them prone to take more risks to relieve the guilt they are experiencing. Possibly, terrorist killers have a paranoid mind-set that can vary from suspiciousness to psychotic delusional thinking. Post70 described the paranoid worldview of political mass murderers such as Hitler and Stalin as well as Ayatollah Khomeini, who mobilized paranoid rage against the “Great Satan” of the United States. The traits in paranoid personalities easily accommodate to the political world because paranoid themes are the leitmotifs in political life. Being suspicious about others who may be seen as not being honest, questioning the loyalty of supporters, constantly being vigilant about information that can be used against them, reading covert meanings in what is said about events, holding grudges, and interpreting comments as attacks that require action and vengeance are part and parcel of the political milieu. Distrusting others and seeing them as hostile while denying these traits in oneself sets the stage for taking homicidal action against one’s enemies in groups committed to violence. More specifically, it would seem that political terrorists are drawn to killing because of a psychological mind-set. The appeal of terrorism is that it allows the perpetrators to commit acts of violence and then rationalize them.75 Their logic is based on a splitting of “us versus them” without any middle ground. Others are seen as the source of evil and harm and the source of their problems; as such, they must be destroyed. There is not only a sense of rage at the “other” but also a logical conclusion that destroying these others is a desirable, if not a necessary and moral, imperative. However, the conclusion is that major psychopathology does not usually exist in the terrorists. Rather, there is a reliance on mechanisms of splitting and externalization as found in borderline and narcissistic individuals to a high degree.

48


WHY HAVE HOMICIDE RATES FALLEN? Historically up until 1965, the homicide rate in the United States was less than 5 per 100,000 population. Over the next 20 years, it increased to a range of 8–10 per 100,000. The homicide rate then declined for a few years, but from 1985 it climbed precipitously, led by juveniles and young adults. This increase was often attributed to the crack epidemic and violence associated with drug distribution. Some of the demographic variables believed to be associated with the increase have been discussed. In the latter part of the 1990s, the rates began to drop to levels between 5.0 and 6.0 in all regions of the United States, with some exceptions in the West. The debate is why this drop occurred. Note that this homicide rate is still far above the rates in most European countries, where the rate has fluctuated between 1.0 and 2.0.44 Theories to explain the drop abound, and they illustrate the diverse influences that have operated.76 Frequently mentioned is the aging population of baby boomers that reached middle age and were beyond the most violence-prone years. The changing scene in many major American cities has also been noted, such as the decrease in barroom killings that were more prevalent in factory towns, especially on payday. The culture of bars itself changed, with more affluent people and females being present. Bar owners always had the power to regulate patrons who drank to excess, but law enforcement began to tighten. Police departments in large cities claimed a major role in lowering the homicide rate because of their more aggressive and imaginative policing. Their approaches varied from the redeployment of police officers to confronting groups of young males loitering about street corners. This approach has sometimes involved community policing, in which a closer relationship between police officers and their precinct populations is fostered. This viewpoint has been criticized for ignoring the effects of increased rates of incarceration.77 Tripling the nation’s prison and jail populations to over 2 million by 2005 is another argument offered on why homicide rates were lowered. The idea is that if people are taken off the streets who might subsequently have committed a murder because of their criminal propensities, the homicide rate will be lowered. By 2004, 10% of all inmates in state and federal prisons were serving life sentences, which cost states about $1 million per inmate serving out the term.78 New gun-control statutes in some jurisdictions may have played some role in reducing homicides, although a countertendency in some states was to pass concealed-gun legislation that could work in the opposite direction.


49

The waning of the crack cocaine epidemic also has been seen as lowering the homicide rate because of the lessened effect on individual users and the reduced drug market, in which young dealers were often armed with semiautomatic weapons. Selling the drug may have seemed like an easy way to make a lot of money, but the ravages of crack became more visible in time. However, in some areas of the country, methamphetamine use has increased among the younger population, along with Ecstasy, to take its place. A factor that has been ignored until recently as contributing to the decline in homicide rates is the improvement in medical care. This may be an offspring of bringing techniques from the battlefield to American cities. Trauma centers and the use of 911 telephone lines or cellular phones are combined with helicopters to evacuate the wounded to a medical facility, much like under combat conditions. A deceiving element lurks in this phenomenon: the number of shootings per se may not be decreasing, but rather, more people are surviving after a shooting. In fact, the number of cases of aggravated assault has not decreased dramatically but remained rather stable. An ingenious and controversial economic theory to explain the decline in homicide rates that occurred by the end of the 1990s has been offered on the basis of the legalization of abortion.79 The explanation offered is that in 1973 the U.S. Supreme Court legalized abortion in Roe v. Wade, whereas abortion had only been available in five states until then. The result was a dramatic increase in abortion numbers from 750,000 to 1.6 million by 1980. The idea is that very often these women were unmarried, adolescents, or poor, in which case the children would have been in a group with the highest potential rates for criminal activity. Thus, in the 1990s the groups that were aborted would have been in their late teens when the rates began to drop. It was noted that the crime rate started falling earlier in the five states that already had legalized abortion and that thereafter the states with the highest abortion rates saw the biggest drop in crime, with other factors being controlled. Lowering of domestic killings needs to be noted as another factor in lowering homicide rates. Society became less tolerant of domestic violence, and many states required police to make an arrest if there was any evidence of physical harm to either party. The Bureau of Justice Statistics data show that since 1976 the number of men murdered by intimates (spouses, ex-spouses, boyfriends and girlfriends) dropped by 68%. The number of women killed by intimates was stalled for two decades but then declined after 1993–1995 and has remained stable into the twenty-first century.


50

It would be unduly optimistic to think that all these variables will remain constant and that no new factors will arise related to homicidal behavior. Any shift in the variables mentioned can upset the recent decline as well. The cyclic nature of homicides over time is what has seemed most consistent in the long run.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14.

15. 16.

17.

Holinger PC: Violent Deaths in the United States. New York, Guilford, 1987 Reiss A Jr, Roth JA: Understanding and Preventing Violence. Washington, DC, National Academy Press, 1993 U.S. Department of Justice, Federal Bureau of Investigation: Uniform Crime Reports. Washington, DC, U.S. Government Printing Office, 1991 U.S. Department of Justice, Federal Bureau of Investigation: Uniform Crime Reports. Washington, DC, U.S. Government Printing Office, 2004 Fox JA, Levin J, Quinet KD: The Will to Kill, 2nd Edition. Boston, MA, Allyn & Bacon, 2005 Tardiff K, Marzuk PM, Leon AC, et al: Homicide in New York City. JAMA 272:43–46, 1994 Wolfgang ME: Patterns in Criminal Homicide. Philadelphia, University of Pennsylvania Press, 1958 Hewitt JD: The victim–offender relationship in convicted homicide cases: 1960–1984. J Crim Justice 16:25–33, 1988 Eisner M: Long-term historical trends in violent crime, in Crime and Justice: A Review of Research, Vol 30. Edited by Tonry M. Chicago, IL, University of Chicago Press, 2003, pp 83–142 Kellermann AL, Mercy JA: Men, women, and murder: gender-specific differences in rates of fatal violence and victimization. J Trauma 33:1–5, 1992 Bartol CR: Criminal Behavior: A Psychosocial Approach, 7th Edition. Upper Saddle River, NJ, Prentice Hall, 2005 Laudanu SF, Drapkin E, Arad S: Homicide victims and offenders: an Israeli study. J Crim Law Criminol 65:390–396, 1974 U.S. Department of Justice, Federal Bureau of Investigation: Uniform Crime Reports. Washington, DC, U.S. Government Printing Office, 2004 Cook PJ: The technology of personal violence, in Crime and Justice: A Review of Research, Vol 14. Edited by Tonry M. Chicago, IL, University of Chicago Press, 1991, pp 1–77 Saltzman LE, Mercy JA, O’Carroll PW, et al: Weapon involvement and injury outcomes in family and intimate assaults. JAMA 267:3043–3047, 1992 Fox JA, Zawitz M: Homicide Trends in the United States. Washington, DC, Bureau of Justice Statistics, 2004. Available at: http://www.ojp.usdoj.gov/ bjs/homicide/homtrnd.htm. Accessed April 28, 2005. Miethe TD, Regoeczi WC, Drass KA, et al (eds): Rethinking Homicide: Exploring the Structure and Process Underlying Deadly Situations. New York, Cambridge University Press, 2004


51

18. Luckenbill DF: Homicide as a situated transaction. Soc Probl 25:176–186, 1977 19. Rosenfeld R, Bray TM, Egley A: Facilitating violence: a comparison of gang-motivated, gang-affiliated, and nongang youth homicides. Journal of Quantitative Criminology 15:495–516, 1999 20. Lipsey MW, Derzon JH: Predictors of violent or serious delinquency in adolescence and early adulthood, in Serious and Violent Juvenile Offenders. Edited by Loeber R, Farrington DP. Thousand Oaks, CA, Sage, 1998, pp 86–105 21. Holcomb WR, Anderson WP: Alcohol and multiple drug abuse in accused murderers. Psychol Rep 52:159–164, 1983 22. Mayfield D: Alcoholism, alcohol intoxication and assaultive behavior. Dis Nerv Syst 37:288–291, 1976 23. Greenberg DF: Methodological issues in survey research on the inhibition of crime. J Crim Law Criminol 72:1094–1108, 1981 24. Cole KE, Fisher G, Cole SS: Women who kill. Arch Gen Psychiatry 19:1–8, 1968 25. Van Hasselt VB, Hersen M: Aggression and Violence: An Introductory Text. Boston, MA, Allyn & Bacon, 2000 26. Roizen J: Epidemiological issues in alcohol-related violence, in Recent Developments in Alcoholism. Edited by Galanter M. New York, Plenum, 1997, pp 7–40 27. Welte JW, Abel EL: Homicide: drinking by the victim. J Stud Alcohol 50:197–201, 1989 28. Lang AR, Goeckner DJ, Adesso VJ, et al: Effects of alcohol on aggression in male social drinkers. J Abnorm Psychol 84:508–518, 1975 29. Gillies H: Murder in the west of Scotland. Br J Psychiatry 111:1087–1094, 1965 30. Derville P, L’Epee P, Lazarin HJ, et al: Statistical indications of a possible relationship between alcoholism and criminality: an inquiry into the Bordeaux region. Revue Alcoholisme 7:20–21, 1961 31. Levi M, Maguire M: Violent crime, in The Oxford Handbook of Criminology, 3rd Edition. Edited by Maguire M, Morgan R, Reiner R. New York, Oxford University Press, 2002, pp 795–843 32. Cloninger CR, Siguardsson S, Gilligan SB, et al: Genetic heterogeneity and the classification of alcoholism, in Alcohol Research From Bench to Bedside. Edited by Gordis E, Tabakoff B, Linnoila M. Binghamton, NY, Hayworth, 1989, pp 3–16 33. Sher KJ, Grekin ER, Williams NA: The development of alcohol use disorders, in Annual Review of Clinical Psychology. Edited by Nolen-Hoeksema S, Cannon TD, Widiger T. Palo Alto, CA, Annual Reviews, 2005, pp 493–523 34. U.S. Secretary of Health and Human Services: Alcohol and Health: Eighth Special Report to the U.S. Congress. Rockville, MD, National Institute of Alcohol Abuse and Alcoholism, U.S. Government Printing Office, 1993 35. Wilsnack RW: Antecedents and consequences of drinking and drinking problems in women: patterns from a U.S. national survey, in Alcohol and Addictive Behavior: Nebraska Symposium on Motivation, 1986. Edited by Clayton P. Lincoln, University of Nebraska Press, 1987, pp 85–158

52


36. Scott CL, Resnick PJ: Assessing risk of violence in psychiatric patients. Psychiatric Times 19:1–16, 2002 37. Pernanen K: Alcohol in Human Violence. New York, Guilford, 1991 38. MacAndrew C, Edgerton RB: Drunken Comportment. Chicago, IL, Aldine, 1969 39. Hanzlick R, Gowitt GT: Cocaine metabolite detection in homicide victims. JAMA 265:760–761, 1991 40. Hagan FE: Introduction to Criminology: Theories, Methods, and Criminal Behavior, 5th Edition. Belmont, CA, Wadsworth/Thomson Learning, 2002 41. Shakespeare W: Othello, in The Complete Signet Classic Shakespeare. Edited by Barnet S. New York, Harcourt Brace Jovanovich, 1972 42. Lieber AL: The Lunar Effect. Garden City, NJ, Anchor Doubleday, 1978 43. Interpol: International Crime Statistics. London, Home Office Press, 2003 44. Barclay G, Tavares C: International Comparisons of Criminal Justice Statistics 2001. London, Home Office Press, 2003 45. D’Emilio J, Freedman EB: Intimate Matters: A History of Sexuality in America. New York, Harper & Row, 1988 46. Maxfield MG: Lifestyles and routine activity theories of crime: empirical studies of victimization, delinquency, and offender decision-making. Journal of Quantitative Criminology 3:275–281, 1987 47. Hindelang MJ, Gottfredson MR, Garofalo J: Victims of Personal Crime: An Empirical Foundation for a Theory of Personal Victimization. Cambridge, MA, Ballinger, 1978 48. Cohen LE, Felson M: Social change and crime rate trends: a routine activities approach. Am Sociol Rev 44:588–608, 1979 49. Carroll L, Jackson PI: Inequality, opportunity, and crime rates in central cities. Criminology 21:178–194, 1983 50. Agnew R: Juvenile Delinquency: Causes and Control. Los Angeles, CA, Roxbury Publishing, 2005 51. Messner SF, Tardiff K: The social ecology of urban homicide: an application of the routine activities approach. Criminology 23:241–267, 1985 52. Merton R: Social structure and anomie. Am Sociol Rev 3:672–682, 1938 53. Perlis RH, Fraguas R, Fava M, et al: Prevalence and clinical correlates of irritability in major depressive disorder: a preliminary report from the Sequenced Treatment Alternatives to Relieve Depression study. J Clin Psychiatry 66:159–166, 2005 54. Agnew RS: Why Do Criminals Offend? Los Angeles, CA, Roxbury Publishing, 2005 55. Sternberg RJ: The Psychology of Hate. Washington, DC, American Psychological Association, 2005 56. Matza D: Delinquency and Drift. New York, Wiley, 1961 57. Binder G: The origins of American felony murder rules. Stanford Law Rev 57:59–208, 2004 58. Dietz ML: Killing for Profit: The Social Organization of Felony Homicide. Chicago, IL, Nelson-Hall, 1983


53

59. Hickey EW: Serial Murderers and Their Victims. Belmont, CA, Wadsworth/Thomson Learning, 2002 60. Johnston T: A little house of horrors: murder in Philadelphia. Newsweek, April 6, 1987, pp 27–28 61. Fox JA, Levin J: Extreme Killing/Understanding Serial and Mass Murder. Thousand Oaks, CA, Sage, 2005 62. Alvarez A, Bachman R: Murder American Style. Belmont, CA, Wadsworth/Thomson Learning, 2003 63. Holmes RM, Holmes ST: Murder in America, 2nd Edition. Thousand Oaks, CA, Sage, 2001 64. Egger S: Serial Murder: An Elusive Phenomenon. New York, Praeger, 1990 65. Fox JA, Levin J, Quinet KD: The Will to Kill, 2nd Edition. Boston, MA, Allyn & Bacon, 2005 66. Godwin GM: Hunting Serial Predators: A Multivariate Classification Approach to Profiling Violent Behavior. Boca Raton, FL, CRC Press, 1999 67. Rutherford A: Crime, Law Enforcement, and Penology. Chicago, IL, Britannica Book of the Year, 1992 68. Malmquist CP: Psychiatric aspects of familicide. Bull Am Acad Psychiatry Law 8:298–304, 1980 69. Proctor RN: Racial Hygiene Under the Nazis. Cambridge, MA, Harvard University Press, 1988 70. Post JM: Assessing leaders at a distance: the political personality profile, in The Psychological Assessment of Political Leaders. Edited by Post JM. Ann Arbor, University of Michigan Press, 2003, pp 69–104 71. Robins RS, Post JM: Political Paranoia/The Psychopolitics of Hatred. New Haven, CT, Yale University Press, 1997 72. Rosenfeld R: Why criminologists should study terrorism. The Criminologist 27:3–4, 2002 73. Crenshaw M: The causes of terrorism, in Violence: A Reader. Edited by Besteman C. New York, New York University Press, 2002, pp 99–117 74. Stern J: Terror in the Name of God: Why Religious Militants Kill. New York, HarperCollins, 2003 75. Post JM: Leaders and Their Followers in a Dangerous World. Ithaca, NY, Cornell University Press, 2004 76. Blumstein A: Disaggregating the violence trends, in The Crime Drop in America. Edited by Blumstein A, Wallman J. New York, Cambridge University Press, 2000, pp 13–44 77. Conklin JE: Why Crime Rates Fell. Boston, MA, Allyn & Bacon, 2003. 78. Butterfield F: Almost 10% of all prisoners are now serving life terms. New York Times, May 12, 2004, p A15 79. Levitt SD, Dubner SJ: Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York, William Morrow, 2005


CHAPTER

2 BIOLOGICAL FACTORS IN HOMICIDE DILEMMAS IN THE BIOLOGICAL PERSPECTIVES ON HOMICIDE A discussion of homicidal violence cannot proceed without first assessing the biology of such extreme aggressive behaviors. The meaning of aggression itself is controversial because some writers confuse it with competitiveness and assertiveness. One approach is to use the term aggression only for behavior that has a component of intended harm. It can be viewed as premeditated, defensive, or impulsive. Another classification distinguishes whether the acts are affective or instrumental in nature. This delineation permits an inquiry into the basis of aggressive behavior and the potential for harm. Although general and developmental psychologists have stressed such factors as the role of frustration and social learning in aggression, the search for the biological source of such behavior has often been given short shrift. Several key questions are relevant when considering biological aspects of aggression related to a homicidal outcome: 1. Is there a common biological substratum for acts of homicidal violence? If so, is homicide just one point, often reached by chance, along a continuum of violent expressions? 2. What is the extent of the current knowledge about the neurophysiological aspects of aggression and how these aspects are related to homicide? 3. How relevant—and how analogous—are the studies of nonhuman aggression and violence to the study of human homicidal behavior? 4. Are there any major limitations to the conclusions that have been drawn about human beings killing one another from naturalistic settings as well as from the artificiality of laboratory settings? Some initial caveats should be stated about the definitions and models of aggression that are used in experiments with nonhuman subjects, be-

55

56


cause such models might be quite different from those applied to humans in a separate milieu. In addition, it is oversimplistic to view nonhuman subjects’ behavior as being programmed in a certain way based solely on genetic underpinnings; sophisticated behavioral geneticists do not espouse such a reductionistic view. Instead, many environmental factors impinge on the behavior of both humans and nonhuman subjects, including their diverse types of social interactions with each other. There is also the role of ritualized and posturing behavior that animals engage in to different degrees. All these considerations need to be factored in before any implications about humans and homicidal violence can be drawn from animal studies. Beyond these asides, in this chapter I do not focus on nonhuman animal displays of aggression or violence in discussing the problem of humans killing. It is commonly believed that biological factors operate in human beings in a dispositional manner—that is, these biological factors are predisposing factors that are seen as being necessary but not always sufficient for a violent outcome of homicidal proportions to occur. As such, these dispositional tendencies stand ready to trigger violent acts framed by environmental events. However, it must be noted that biological fluctuations within individuals can also heighten or lessen the potential for such behavior to be triggered at different times.

EARLY THEORIES OF THE SOURCE OF AGGRESSION Instinctual and Behavioral Theories The search for an instinctual basis for aggression has an uneven history. In the early twentieth century, McDougall1 postulated that humans possess a host of instincts that operate for acquisition, construction, gregariousness, flight, repulsion, curiosity, pugnaciousness, parenting, reproduction, desire for food, self-abasement, and self-assertion. Eventually, however, the theories of behaviorism became the reigning school of psychology in the United States through the 1950s. Believing that scientific validity could be proven only by observable events, behaviorists found McDougall’s unprovable theory about an inner set of processes driving behavior unacceptable and replaced it with a drive theory based on external observable behavior, without reference to the inner state of a person. The emphasis on behavior as being primarily learned dominated psychological thinking at the time; it was felt that the idea of behavior being unlearned and inborn was not testable and as such was therefore not a legitimate part of the contemporary scientific domain. Environmental contributions were seen as always controlling behavior, beyond some basic motor patterns for reflexes.

Biological Factors in Homicide

57

Psychoanalytic Theories Psychoanalysts have formulated general theories on the concept of instinct, focusing on questions such as the number and nature of the instincts, how they should be classified (e.g., energic aspects, libido theory), how they function, and how they could be transformed (e.g., instinctual vicissitudes). A special theory of instincts referred to how instincts develop in any one particular person. Throughout the twentieth century, these theories shifted. Underlying the various theories was the idea that some built-in source for behavior was distinct from the aim of the instincts or the objects on which they were played out. Originally, instincts were divided into the sexual and ego instincts. Freud later ascribed aggressive trends to the ego instincts.2 With the addition of structural theory, aggressivity was no longer regarded as primarily attributed to the ego instincts. Rather, aggression and destruction were postulated as being independent instincts alongside the sexual instincts. It is beyond the scope of this chapter to go into Freud’s speculations about a death instinct based on his need to explain how some behavior was not regulated by the pleasure principle but rather operated under the sway of a repetition compulsion.

Ethological Theories In an entirely different arena of knowledge, European zoologists were laying the groundwork for studying instincts by way of ethology. Their methodology was to investigate animal behavior within animals’ naturalistic settings. From studies of lower animals, it had become difficult to explain how all behavior operated simply on the basis of being learned. This is not to say that such behavior was immune from environmental influences, but rather that the genetic endowment of the animal was seen as interacting with the environment. The ethologists’ reasoning was that instinctive behavior was genetically programmed and therefore related to inheritable characteristics. Rather than referring to instincts, ethologists referred to innate behaviors that were specific to a given species. At times, confusion has resulted from the misrepresentation of the position that the phrase “inherited behavior” simply means that a group of genes can explain all behavior. This reductionistic view of genetic endowment as an explanation of how aggression unfolds—up through acts of killing—omits too many key variables. Rather, the interaction of both genetic factors (the dispositional component) and environmental factors (the precipitating component) is needed for reaching the potentiality of a homicide.

58


Much of Lorenz’s3 and Tinbergen’s4 early work, which involved concepts such as a “sign stimulus” and “fixed-action patterns,” has been elaborated on by other investigators. More current research has shown that complex, inborn behavioral patterns are activated by certain environmental stimuli that thereby function as releasers. Different species have different repertoires of fixed-action patterns. In lower species, behavior is released by stimulation of certain command neurons, whereas in mammals it is thought that groups of cells may trigger preprogrammed motor acts. Command systems in turn may be responsive to certain sensory elicitors by innate releasing mechanisms. Although most of the studies of these concepts have not been based on work with human subjects, let alone even mammals, they have provided a framework for thinking about the biological contributions to aggression in its diverse forms. Seeing the influence of genes on behavior sparked the transition from a behavioristic model to one that incorporates an impetus from within the organism. If behavior is regulated by neural circuits, then diverse proteins are required for the development of these circuits. The proteins are coded specifically by genes. Different genes may play a role, with some individual genes becoming more important than others. In essence, the genes are the key to producing the neural circuitry and also to regulating behavior. Genes code the structural proteins necessary to maintain the circuitry as well as the enzymes necessary for synaptic transmission. Genes are also seen as coding peptide hormones and modulations that can trigger or inhibit the discharge of behavior. (Peptides are neurotransmitters that affect specific neurons; these are discussed later in this chapter’s section “Role of Neurotransmitters.”) The complex question of what role innate factors play in determining human behavior remains. The genetic studies that have been done on humans have not found a definitive association among impulsivity, aggression, and reduced serotonin behavior, although rats can be bred for aggressive behavior.5 However, besides studies on the hormonal determinants of gender identity, five lines of evidence are presented in support of how some human behaviors reflect innate factors:6 1. Human behavior, like that in other species, is mediated by genes. Consequently, to some extent, it is under genetic control. 2. Certain emotional expressions are seen as having a strong innate component that cuts across different cultural settings. An example would be that of facial expressions in infants or adults, regardless of those individuals’ home culture. 3. Stereotyped motor patterns exist that resemble fixed-action patterns.


59

4. Certain complex patterns appear to require little or no learning. This is witnessed in humans who have had limited experiences of the environment or in the responses of blind infants who have not had any visual environmental input. 5. The development of language in humans, once viewed as simply learned, is believed to have a major biological predisposition. Children in diverse parts of the world seem to display similar linguistic achievements at the same age. Linguistic universals suggest that language acquisition is biologically programmed. It develops based on the child’s biological predisposition, along with cognitive development and the linguistic environment. In the remainder of the chapter, I focus on the biological—including genetic—factors that influence human aggression.

GENES AND VIOLENCE IN HUMANS Genetic factors are being investigated for their role in diverse kinds of human behavior that have an influence on how aggression may be expressed. One difficulty in such work is separating the role of genetic influences from that of environmental influences. Much of the genetic work in humans stems from the field of molecular genetics, in which techniques are used to localize, characterize, and clone genes believed to be associated with inherited diseases. By use of an oversimplified model, the steps used by molecular geneticists are as follows: First, they locate a precise gene that can determine a genotype (i.e., the genetic constitution of an individual or group). Then they determine the products of a specific defective gene (e.g., alterations of some biochemical product) to reveal the biological mechanisms involved. The hope is that through this process, the underlying pathophysiology of a certain disorder or condition will become clearer and assist investigators in reaching more valid clinical diagnoses. Genes for certain neuropsychiatric disorders (e.g., Huntington’s disease) and certain metabolic diseases are typically cited as evidence for inherited diseases. It is hoped that these DNA methods can be extended to the study of mental illnesses and therefore yield a better understanding of the violent propensities that may accompany some of these illnesses. Traditional approaches to studying genetic transmission have focused on the incidence and prevalence of disorders in families, twins, and adopted children, with the goal of separating genetic from environmental influences to the extent possible. Focusing on the similarities and differ-

60


ences in groups that have a similar genetic makeup, this traditional model has been extended beyond the search for causes of mental disorders to a search for those genes that simply express inherited traits. For example, if studies of monozygotic twins with certain traits who were separated at birth were later to reveal that both twins had an increased potential for violence in adulthood, perhaps even something as specific as homicidal violence, questions could be raised about the possibility of genetic influences on violent behavior. In the same vein, if one dizygotic twin separated at birth showed a comparatively lesser incidence of such behavior than the matching dizygotic twin, the same possibility would exist. Studies of chromosomal abnormalities, such as the “supermale” XYY, were another approach, studying whether the extra Y chromosome would be related to increased violence, such as seen in Richard Speck. Speck murdered eight nurses and was an XYY male; however, studies using random populations have failed to confirm the XYY–violence hypothesis.7 Family studies are the weakest line of evidence for genetic transmission because of the obvious loading of similar environmental variables in family rearing patterns. Unfortunately, they are the most prevalent model used for studies of mental disorders. For example, different studies on twins and adopted siblings point to genetic factors having some role in schizophrenia and certain mood disorders. In this connection, using adoption studies, investigators look not only at the mood instabilities present in those with an affective disorder but also at these individuals’ violence potential in comparative studies. As another example, attempts might be made to connect the hereditary components in alcoholism with mood instabilities and to go beyond that into the summative disposition to violence. Research with monozygotic twins indicates a hereditary aspect to aggressive behavior because concordance rates for monozygotic twins were greater than for dizygotic twins.8 Twin studies suggest antisocial behavior in adults may be more related to genetics than to environment. The complexities of investigating genetic components are increasing. For many diagnoses or conditions, researchers are still generating hypotheses, such as a single gene locus versus a multifactorial model involving polygenes in which environmental influences operate. At a molecular level, hypotheses involve the role of neurotransmitter enzymes and metabolites and how they can affect neurophysiological functioning. Even for diagnosable mental disorders or biological traits, such as schizophrenia or mood disorders, the results remain inconclusive. This ambiguity is even greater with traits related to violent behavior. The statistical models do not fit the data to the exclusion of other hypotheses.9 The missing link would be data to show the coinheritance of an illness and a biological trait.


61

In studies of genetic aspects of mental disorders and related traits, the current emphasis is on markers or genetic variations, with the goal being to localize genes. The hope is that molecular techniques can bridge the distance between a marker and a gene by identifying additional markers. Such work forms the basis for reports of linkage between diagnoses of bipolar affective disorder and genetic markers on the X chromosome, which suggests one form of a genetic bipolar illness. The complexities of the approach reveal obscure inheritance patterns and the fluid boundaries existing for many diagnoses. The increasing amount of genetic knowledge about mental disorders raises diverse problems, particularly in the areas of social control and criminal responsibility.10 Questions of determinism and responsibility have been argued endlessly in philosophy. Genetic information may make the public more skeptical of the degrees of freedom an individual has in homicidal behavior. Attempts to use a “genetic defense” in homicide cases may well increase, albeit the legal system may insist that the legal rules do not change for responsibility and punishment, regardless of scientific knowledge. Although there is now increasing knowledge from the twin and family studies about a genetic component in major mental disorders, such as schizophrenia and bipolar disorder, and perhaps in antisocial personality disorder as well, debate continues on how genetic influences operate with respect to violence. Epidemiological studies may suggest an increased incidence of violence compared with a control group of those without a mental disability, but the majority of persons with mental disorders still do not commit acts of serious violence. Research also does not allow us to make such predictions based on an increased predisposition. The difference ultimately may lie in whether juries believe there is now hard scientific evidence that in the absence of a genetic vulnerability, the violence would likely not have occurred. However, this perception of evidence may simply produce arguments that genetic predispositions simply make it more difficult to control one’s impulses, a similar view taken for those persons coming from an adverse social environment. Even if cognitive functioning is adversely affected, it may not affect legal responsibility. The U.S. Supreme Court upheld a statute in a Montana case, in which the jury was instructed that it could not consider an intoxicated condition “in determining the existence of a mental state which is an element of the offense.”11 The failed effort of the defense attempted to show evidence of intoxication negating “purpose” or “knowledge” of the homicide. In an earlier case, the XYY chromosome abnormality was tried as a legal defense, in which it was argued that the abnormality in-

62


creased the disposition toward violence. The ruling held that the genetic imbalance must “interfere substantially with the defendant’s cognitive capacity or his ability to understand the basic moral code of his society.”12 Genetic vulnerability is thus handled legally analogous to people who have sustained environmental insults and who thus are more prone to violence—such as persons exposed to serious abuse as children when there was no capacity to influence their damaging childhood; such individuals are held to the same standard of responsibility as other adults. Of course, genetic, as well as environmental, factors may play a role at the sentencing stage after responsibility is established.

PHYSIOLOGICAL HYPOTHESES OF VIOLENCE Because genetic theories of violence currently operate primarily at the hypothesis stage, physiology is often suggested as a more concrete explanation for aggression that can lead to the extreme of killing.

Seizures Work with clinical populations in cases involving seizure disorders when a subgroup has been violent suggests one area for exploration. Early researchers speculated about the possibility of violent eruptions occurring in connection with temporal lobe seizures, positing some type of brain abnormality on an anatomical or physiological level. In the modern era, the search has turned to work on brain functioning for an explanation. In the 1970s, some investigators studying a form of seizure now called complex partial seizures (also called temporal lobe, limbic, or psychomotor seizures) began to advocate the use of neurosurgical techniques to try to control such seizure activity.13 The inference was that if the seizures could be controlled, then associated sequelae of the seizures, such as violence, could also be controlled. However, surgical interventions of this type could not be carried out on large numbers of people without eliciting a large public outcry. Something inherent in the performance of surgical procedures on the brain for the purpose of altering behavior arouses both public and professional anxiety. Part of this concern is associated with the fact that psychosurgery procedures on mentally ill individuals were less well regulated in the past. People also fear that permitting surgery of this type would allow future abuse of the procedure in attempts to control people by overdiagnosing com-


63

plex partial seizures in those prone to violence. The difficulty in diagnosing such seizures because of unclear boundaries in the diagnosis also elicits fears that such a technique could be misused. That neurologists have been able to delineate diverse types of seizures, each of which might have different correlates for different types of violence—perhaps including homicidal violence—raises many potential diagnostic problems but also has implications for assessing violent individuals’ behavior. In Table 2–1, the diverse types of seizures and their characteristics are represented, such as whether they are generalized or partial, differences connected with age at onset, if they are primary or secondary, and the diverse settings in which they can occur. Generalized seizures are those that begin bilaterally; partial seizures have a localized onset, although they may evolve into generalized seizures of the partial complex type (here partial refers to consciousness being retained and complex to an impairment in consciousness). The potential for ambiguity in distinguishing violence due to seizures from associated psychiatric phenomena is evident, and this is magnified when there are legal issues.

Attempts to Link Temporal Lobe Seizures With Violence Early theorizing about temporal lobe seizures was partially correct in that the electrical discharges for these seizures have been found to originate in deep, medially placed limbic nuclei in the temporal lobe, consisting of the amygdala, uncus, and hippocampus. However, the limbic system is not all located in the temporal lobe, and other areas of the brain also affect emotion. It is interesting that some view the limbic system more as a philosophical concept than an actual discrete anatomical or physiological system.14 The epidemiological question of whether people with temporal lobe seizure disorders actually have an increased tendency toward violent behavior has not been satisfactorily resolved. Many times such a link has been implied in best-selling books or movies, such as Michael Crichton’s novel The Terminal Man, in which seizures spark brutal rampages.15 However, authors of more recent studies did not find extreme violence occurring more frequently in the course of complex seizures.16,17 The difficulty is in the reflexive acts, which do not have an intentional component, and are a set of behaviors carried out without any particular goal or purpose in mind. To further complicate the matter, some neurologists have posited that people with temporal lobe seizures may develop behavioral changes involving excess aggression, religiosity, changes in sexual behavior, and a preoccupation with writing or drawing (which some also think promotes creativity).


64 TABLE 2–1.

Classification of epileptic seizures

Generalized seizures (bilaterally symmetrical and without local onset) Tonic, clonic, or tonic-clonic (grand mal) Absence (petit mal) Simple—loss of consciousness only Complex—with brief tonic, clonic, or automatic movements Lennox-Gastaut syndrome Juvenile myoclonic epilepsy Infantile spasms (West syndrome) Atonic (astatic, akinetic) seizures (sometimes with myoclonic jerks) Partial, or focal, seizures (seizures beginning locally) Simple (without loss of consciousness) Motor (tonic, clonic, tonic-clonic; Jacksonian benign childhood epilepsy; epilepsia partialis continua) Somatosensory or special sensory (visual, auditory, olfactory, gustatory, vertiginous) Autonomic Psychic Complex Beginning as simple partial seizures and progressing to impairment of consciousness) With impairment of consciousness at onset Special epileptic syndromes Myoclonus and myoclonic seizures Reflex epilepsy Acquired aphasic and convulsive disorder Febrile and other seizures of infancy and childhood Hysterical seizures Source. From “Proposal for Revised Clinical and Electroencephalographic Classification of Epileptic Seizures. From the Commission on Classification and Terminology of the International League Against Epilepsy.” Epilepsia 22:489, 1981.

A different conceptualization interprets the problem not just as a seizure disorder but rather as a dysfunction in the limbic system that is clinically referred to as episodic dyscontrol.18 Neural circuits related to defensive rage behavior respond to inescapable threats and activate the periaqueductal gray of the midbrain and its projections. 19 In turn, limbic and cortical influences on the periaqueductal gray may be direct or may be funneled through the hypothalamus. Activation of the medial hypothalamus or amygdala then facilitates defensive rage. Another approach has been to focus on seizure-prone individuals’ interictal personality. It has been proposed that these individuals are more likely to be prone to violence between seizures rather than in sei-


65

zure states and that such a predisposed personality shows traits of irritability and aggressivity that leave those individuals prone to having violent outbursts or to provoking others. Difficulties thus abound in trying to achieve specificity in the association between complex partial seizures and violence, just as with various psychiatric disorders and aggression.20,21 Some investigators have classified patients as showing increased aggressive behavior. However, in these studies, varying definitions of aggression have been used, and in many of these cases these definitions fall far short of any type of serious violence toward people, let alone any homicidal violence. Whitman and colleagues22 studied 83 children who were diagnosed with anterior temporal lobe epileptiform spike activity. The children showed elevated aggression scores on the Child Behavior Checklist.23 Many variables had only a slight variance to predict behavioral disorder for generalized and partial seizures. The authors eventually focused on situation-centered variables instead of biological variables to try to predict behavioral disturbances. In the early 1980s, Lewis and colleagues24 investigated psychomotor seizures in a group of 97 young offenders. These authors believed that there was an association between the offenders’ seizure disorders and violence. Eleven of the youths were seen as definitely having psychomotor seizures, and 8 were thought to probably have such a diagnosis. The 11 had been seriously assaultive, with 5 described as having committed acts of violence during the seizures. Complicating the clinical picture, however, was the frequency of antecedent central nervous system trauma in the backgrounds of these youths. The group showed a loading of factors, such as complicated perinatal problems, head injuries, central nervous system infections, and histories of some generalized seizures of the grand mal type. Some also had psychotic symptoms as well as a history of severe physical abuse. Given such a matrix of factors, which is often found in such clinical populations, attributing these youths’ increased violence to complex partial seizures is like looking for a needle in a haystack. In another study of a possible relationship between violence and seizures, Hermann and Whitman25 did not find any significant differences in the occurrence of violence between those who did and those who did not have seizures. Study subjects who had seizures were weighted on such variables as socioeconomic status, gender, age, and earlier developmental problems, much as in other studies. It was found that apart from their seizures, these individuals had multiple problems, some of which may have originated in the subjects’ deprived social environment and may have been connected with diagnoses of attention-deficit

66


disorder with or without hyperactivity. Their problems took the form of coexisting restlessness, impulsivity, poor attention span, memory deficits, and so on. Some subjects’ history of accident proneness or head injuries may also have left sequelae. Interpretations of the data were complicated by the study subjects’ learning disabilities, soft neurological signs, and a variety of neuropsychiatric deficits that did not fit any one standard accepted diagnosis and may have included other diagnoses as well. It is extremely difficult to sort out diverse factors such as those found in Herman and Whitman’s study from what may be directly connected with the effects of seizures. The brain is not one discrete area but instead encompasses the cortical area of the hippocampus, the frontalorbital area of the cortex, and the cingulate gyrus. There is a high degree of connection between these limbic areas and the temporal lobes, and any hypothesized lesion need not occur in any one area. In fact, the seizure disorder is often ascribed to existing on a physiological basis leading to a dysfunction that is not specifically localized in any one anatomical area. Many would also question the sensitivity of diagnostic instruments and practices to detect what might be associated with limbic seizures. Electroencephalograms (EEGs) are seen as too insensitive to detect abnormalities in subcortical activity, and neuropsychological testing cannot be relied on to differentiate such a diagnosis. Clinical practice relies on the old-fashioned diagnostic approach of history taking and trials of antiseizure medications in the absence of specific ways to validate the diagnosis. If the problems improve with medication, then the diagnosis is retroactively seen as being confirmed. Thus conflicting views have emerged regarding the proneness of those individuals with seizure disorders toward exhibiting extreme violence. At this time, it cannot be confirmed or disconfirmed in principle that these individuals have such a vulnerability, except for citation of specific cases.

Possible Linking Mechanisms Between Seizures and Violent Behavior Supposing for a moment that there is a connection between seizures and violence, how might a homicidal act occur in an individual during a seizure? The primary manifestations of a complex partial seizure begin with an alteration in the individual’s state of consciousness, specifically described as a temporal aura spreading to the limbic system. Stages of impaired consciousness then occur, with degrees of loss of contact within the environment. Sometimes a focal onset progresses to a loss of con-


67

sciousness, or at times a loss of consciousness exists from the beginning. Automatisms may occur that appear quite integrated to an observer but for which the individual has postictal amnesia. The hypothesis is that during automatist states a violent act would occur. Possible automatisms are those involving movements or activities (e.g., chewing, scratching, walking, kicking, undressing) and those described as being responsive to stimuli in the environment (e.g., drinking, resisting movement by others such as when being placed in restraints). In addition, whatever behavior was occurring before the seizure, such as eating or walking, may be continued. These diverse behaviors occurring before and during a complex seizure make it difficult to attribute an act of aggression or violence to the complex seizure. In addition, automatisms sometimes occur after a seizure and at other times in the postictal state. Attributing causation for violence to the seizure is difficult because one would have to reconstruct the perpetrator’s neuropsychological state at the time of the act without the benefit of contemporaneous EEGs or other measurements. Interictal behavior proposed to be associated with an increased tendency to violence, assuming such a correlation has validity, may also be attributable to several possibilities: 1) the personality of the person prone to such seizures, 2) personality changes in reaction to being seizure prone, or 3) a hypothesized, undetected focus. Again, none of these possibilities can be confirmed or disconfirmed. Some hope that the ambiguity may be resolved by stereo EEGs, which involve surgical insertion of wire electrodes into the brain to monitor brain abnormalities. However, this interesting research approach is not likely to attract many volunteers. Psychiatric signs and symptoms connected with the seizures, whatever their cause, provide further complications. Autonomic nervous system involvement can show up in gastrointestinal, cardiovascular, pulmonary, or urogenital symptoms. Affective states of depression, pessimism, anxiety, or euphoria may prevail. A picture of confusion with memory alterations may impair the later recall of details. In some individuals, complex partial seizures evolve to a state of generalized seizures. Repeated seizures in themselves can induce brain injury and associated behavioral changes. Occasional electrical discharges or degrees of undetected discharge could be occurring. From a psychological perspective, the tendency to violence could be explained by the individual’s reaction to having seizures, which are, after all, a disease of the central nervous system. Often, a seizure-prone individual may have a comorbid diagnosis that involves schizophreniform or paranoid-type personality features that are reflected in mood disturbances. The personality features most frequently referred to are hyper-

68


religiousness, hyposexuality, hypergraphia, and circumstantiality. Auditory hallucinations and paranoid delusions can also occur. There is a need to determine whether these symptoms are related to schizophrenia. To complicate matters further, the onset of psychotic symptoms may occur years after the onset of the seizures. In one study of 44 epileptic patients, 20 met the criteria for a schizophrenic disorder.26 Most elusive is the attempt to portray a “dyscontrol syndrome” as producing violence from lesions in the limbic system or temporal lobe in the absence of any observable seizures. This hypothesis suggests that there are epileptic types of discharges from the limbic system, based on suggestive evidence from soft neurological signs, electroencephalographic abnormalities, attention-deficit disorders, and so on. This vague group of signs, symptoms, and diverse diagnostic possibilities exhibits diverse types of dyscontrol in terms of frequency and severity. When used in attempts to explain homicidal behavior, this grouping thus appears to be elusive, at present not capable of being falsified as a diagnosis from its elastic boundaries. It is likely that greater diagnostic specificity will be obtained in the future from electroencephalographic work along with more advanced brain imaging techniques. Although large numbers of electroencephalographic studies have attempted to detect significant brain abnormalities in individuals ranging from delinquents and psychopaths to murderers, major problems exist in interpreting these reports.27 The methodological difficulties involve lack of control subjects, the subjective nature of scoring EEGs, lack of blind assessment of the records, lack of agreed-on criteria for defining abnormality, and lack of sensitivity to states of alertness of the subject. Perhaps the newer brain imaging techniques will allow greater specificity, such as pinpointing temporal lobe abnormalities through computed tomography. In one study of 87 patients with diverse psychiatric diagnoses who were referred for neuropsychiatric evaluation, both computed tomography and magnetic resonance imaging were performed.28 Twenty-three of the patients were diagnosed as having an organic mental disorder; significantly, the 14 of those 23 who had a history of frequent violent episodes showed lesions in the anterior-inferior temporal lobe, whereas the 9 who were not violent did not show such lesions. In summary, once a seizure-prone individual has carried out a homicide or serious act of violence, several possibilities of the biological cause arise.29 One possibility is that the act might have been caused by the seizure, although such violence as part of an ictal state is quite rare. Another possibility is that a type of cerebral malfunction may be the cause of both the homicide and the seizure disorders. Sometimes the violence may be associated with the confusion of a postictal state. Yet another possibility


69

is that some types of psychological sequelae accompanying a seizure problem may foster resentment and accumulated rage, leaving the person prone to a homicidal outburst, or a mental disorder may have developed as a result of the seizure problem and may have played some contributory role in the homicide. Also to be considered is the background of psychosocial deprivation that can contribute to an individual’s seizure proneness as well as to homicidal behavior. Furthermore, in some of these cases, patterns of violent or provocative behavior, whatever their source, can elicit reactive trauma and the possibility of resultant brain injury from such encounters.

Social and Legal Issues About Seizures and Violent Behavior The potential connection between complex partial seizures and homicidal violence raises complicated social and legal issues beyond those of proper medical diagnoses. The essentials for a criminal act are an actus reus (criminal act) as well as a mens rea (criminal mind or criminal intent). The latter involves the idea of an individual voluntarily choosing to commit a criminal act, in which case his or her behavior is seen as blameworthy. In the absence of such a choice, blaming and punishment would be seen as not only inappropriate but unjust. The possibility of a criminal act occurring in an ictal state raises questions about the perpetrator’s volitional capacity. A more subtle and frequent consideration is whether an individual can be cleared of responsibility for interictal acts of violence, with the act being attributed to the “epileptic personality.” However, in legal settings, questions are not posed in epidemiological terms. Instead, they involve the question of whether, in a particular case, an individual can validate the diagnosis of an epileptic personality and whether the homicidal behavior can be encompassed under the seizure diagnosis. Then, depending on the jurisdiction, different types of legal questions arise with respect to an individual’s capacity to control his or her violent behavior, conform his or her conduct to the requirements of the law, know the difference between right and wrong, and know the nature of the act carried out. A legal distinction is made between an automatism defense and an insanity defense. The former is raised by way of a denial of homicidal responsibility because the actus reus is required to have been voluntary. A homicide occurring during an alleged seizure state could raise the question of whether the mens rea was negated because the perpetrator did not register the requisite cognitive or emotional states. A pure automatism defense would not necessitate proving the presence of a mental disease or defect. If successful, the defense would have a “not guilty”


70

outcome in contrast to a “not guilty by reason of insanity” outcome. Hence, the automatism defense would bypass the types of questions raised in using an insanity defense and the types of restrictions normally associated with a successful insanity defense, such as medical or court supervision. One interesting question would be the possibility of a later recurrence of the seizure and potential for violence in the lack of such supervision. Complex legal as well as psychiatric issues arise concerning assertions that voluntariness or its absence in conjunction with a seizure disorder exculpates a defendant for a homicidal act. It seems clear that simply making an assertion of a connection between a seizure disorder and involuntary behavior does not suffice for legal or medical purposes to establish a connection to the act of violence. Furthermore, if medical documentation exists about such a seizure disorder, it may become legally significant when an issue of responsibility is raised. The need to validate the diagnosis by a competent neurologist experienced in the area of seizure disorders is only the beginning. If the accused person is a minor, the diagnostician should be someone who is also experienced in and conversant with the diagnosis in that age group. The question of automatisms needs to be established with electroencephalographic telemetry, and some would add that the testing should occur on closed-circuit television as well. Demonstrations of aggression related to a seizure should be handled by an attempt to induce the seizure and have it on videotape with accompanying electroencephalographic readings. Even if the results were positive, they could still be challenged on the basis that the automatisms were not what actually occurred during the commission of the violent act in question. However, the results would support the claim that analogous behavior did occur. The individual’s history should also indicate past episodes of aggression or violent behavior. For legal purposes, a neurologist or psychiatrist would have to be willing to express an opinion that these factors were related to the seizure disorder and be available for answering other legal questions.

Case Example30 Seth Grant was one among several individuals in a bar. A patron, whom he did not know, and the tavern owner got into an altercation. When the police arrived, the patron was escorted outside. A hostile crowd of about 40 people was cheering the patron when suddenly, Grant, heretofore totally uninvolved, leaped into the air and struck an officer twice in the face. Great force was needed to subdue Grant, who was described as excited, agitated, and upset. In jail 1 hour later, he was observed having symptoms of a grand mal seizure.


71

Grant’s medical history revealed a preexistent diagnosis of psychomotor epilepsy for which he was receiving medication and also verified that he had had a number of violent attacks of varying severity. A previous attack in which Grant assaulted someone with a knife in a hospital had resulted in his being shot in the pelvis and kidney by a police officer. At his current trial, Grant testified that he recalled nothing about the bar episode and had only “woken up” in the hospital 3 days later. His psychiatric expert testified that he had had a psychomotor seizure that “prevented his conscious mind from controlling his actions.” There was conflicting testimony about his being “alert and awake” at the time of the arrest and in jail. A question was also raised about the grand mal seizure in jail, because it would not necessarily mean that he had had a psychomotor seizure while at the bar. The jury was instructed in terms of the Illinois Test for Insanity, which stated that as a result of a mental disease or defect, the defendant would have to lack substantial capacity to either appreciate the criminality of his conduct or conform his conduct to the requirements of the law in order to be found not guilty by reason of insanity. The jury found Grant guilty of aggravated battery and obstructing a police officer, but the court sentenced him only on the former charge. On appeal, it was noted that the insanity question did not distinguish an individual whose behavior was caused by an automatism during which the defendant was not conscious of what he was doing. No instruction had been given the jury involving involuntary conduct, and on that ground the case was reversed and remanded with directions for a new trial.

Apart from other issues, this case example raises the interesting legal dilemma of what instructions should be given in such cases. Cases involving voluntary conduct cause difficulty not only for philosophers and clinicians but also for the legal system, with its need to assess culpability on the basis of an individual performing a voluntary act. The Model Penal Code, which produces recommendations by a scholarly group of attorneys and law professors, lists exclusions from the definition of a voluntary act:31 1) a reflex or convulsion, 2) a bodily movement during unconsciousness or sleep, 3) conduct during hypnosis or resulting from hypnotic suggestion, or 4) a bodily movement that is not a product of the effort or determination of the actor, either conscious or habitual. This scholarly recommendation states that a voluntary act should have an element of conscious control over the behavior but does not attempt to resolve the age-old debate between free will and determinism or how much control. The code is an attempt to exclude, in a crude way, behavior seen as nonvolitional. The rationale on one hand is that a person’s sense of autonomy would be undermined if he or she were held accountable for behavior that he or she could not control. It is also argued that a sense of justice is vitiated in holding such

72


people accountable. The problem arises in a broader sense from many people thinking they cannot control themselves. Hence, as a minimum, there is a requirement for some attendant alteration of consciousness. Of course, because an unconscious person cannot carry out any act, there remains a quandary that produces the doctrine of automatism, implying some (but how much?) limitation of the awareness of his or her acts.

Altered States of Consciousness The list of clinical conditions in which automatisms are alleged to occur has been widely expanded from its origins to now include possible problems on a genetic, hormonal, or neurochemical level. Common law cases involving alleged sleepwalking go back to the seventeenth century in England, although it was not until the nineteenth century that juries began to consider such cases as “fractured or suspended consciousness.” Medical texts began to describe such states as “doubly conscious.”32 The idea emerged that some type of suspension of full human agency was present although not delusional at the time of a homicide. The range of diverse conditions tend to blend in with the dissociative disorder group in which no organic pathology is being asserted. This change increases the legal confusion, because it increases the range of medical experts who are willing to testify on opposite sides and render diverse opinions. The possibilities then include seizures, postictal states, and a variety of altered states of consciousness that can either accompany delirium and dementia or exist in their own right. Altered states can also occur from infectious processes, toxins, tumors, head injuries, and various metabolic diseases such as diabetes or Cushing’s syndrome. Dissociative states, whatever the etiology, pose challenging legal questions. Should people who commit homicidal acts while in such states (assuming clinical evaluations support the existence of such states) be held criminally liable to the same extent as someone without such states? Do these mental states provide a basis for a mental illness defense? If not, should the individuals be punished to the same degree as someone committing such acts but without an altered state of consciousness? Questions about hypnotized individuals carrying out criminal acts continue to be argued without a satisfactory resolution. The affirmative argument against liability for these individuals’ behavior is that they are subject to, and dependent on, the hypnotizer. Yet ascribing such helplessness to the individual does not square with current knowledge


73

about the nature of hypnotizability, in that some may choose to submit to such procedures and others may be extremely suggestible by nature. Matters become even more debatable when automatic states related to somnambulistic (i.e., sleepwalking) or hypnagogic states are suggested. Somnambulistic states always raise interesting legal questions about the nature of responsibility for behavior. Excerpts from the King v. Cogdon case illustrate fundamental issues about responsibility for criminal behavior in this regard.33 This unreported case, heard in the Supreme Court of Victoria before Mr. Justice Smith in December 1950, although clear as to its facts and unchallengeable in law, was seen as compelling a reconsideration of some of the basic premises of responsibility for criminal actions.

Case Example Mrs. Cogdon was charged with the murder of Pat, her only child, a 19year-old woman. Pat had for some time been receiving psychiatric treatment for a relatively minor neurotic condition of which, in her psychiatrist’s opinion, she had been cured. Despite this remission, Mrs. Cogdon continued to worry unduly about her daughter. Describing the relationship between Pat and her mother, Mr. Cogdon testified: “I don’t think a mother could have thought any more of her daughter. I think she absolutely adored her.” On the conscious level, at least, there was no reason to doubt Mrs. Cogdon’s deep attachment to her daughter. To the charge of murdering Pat, Mrs. Cogdon pleaded not guilty. Her story, although somewhat bizarre, was not seriously challenged by the Crown and led to her acquittal. She told how on the night before her daughter’s death, she had dreamed that their house was full of spiders and that these spiders were crawling all over Pat. In her sleep, Mrs. Cogdon left the bed she shared with her husband, went into Pat’s room, and awakened to find herself violently brushing at Pat’s face, presumably to remove the spiders. This woke Pat. Mrs. Cogdon told her she was just tucking her in. At the trial, Mrs. Cogdon testified that she had been told and still believed that the occupants of a nearby house bred spiders as a hobby, preparing nests for them behind pictures on their walls. It was these spiders that in her dreams had invaded their home and attacked Pat. Mrs. Cogdon had also had a previous dream in which ghosts sat at the end of her bed, and she said to them, “Well, you have come to take Pattie.” It did not seem fanciful to accept the psychological explanations of these spiders and ghosts as the projections of Mrs. Cogdon’s subconscious hostility toward her daughter, a hostility that was itself rooted in Mrs. Cogdon’s own early life and marital relationship. The morning after the spider dream, Mrs. Cogdon told her doctor about it. He gave her a sedative and, because of the dream and certain previous difficulties she had reported, discussed the possibility of psychiatric treatment. That evening, Mrs. Cogdon suggested to her husband that he attend his lodge meeting and asked Pat to come with her

74

HOMICIDE: A PSYCHIATRIC PERSPECTIVE to the cinema. After he had gone, Pat looked through the paper, found no tolerable program, and said that because she was going out the next evening, she thought she would rather go to bed early. Later, while Pat was having a bath preparatory to retiring, Mrs. Cogdon went into her room, put a hot water bottle in the bed, turned back the bedclothes, and placed a glass of hot milk beside the bed ready for Pat. She then went to bed herself. There was some desultory conversation between them, and just before she put out her light, Pat called to her mother, “Mum, don’t be so silly worrying there about the war—it’s not on our front doorstep yet.” Mrs. Cogdon went to sleep. She dreamed that “the war was all around the house,” that soldiers were in Pat’s room, and that one soldier was on the bed attacking Pat. This was all of the dream she could later recapture. Her first “waking” memory was of running from Pat’s room, out of the house to the home of her sister, who lived next door. When her sister opened the front door, Mrs. Cogdon fell against her, crying, “I think I’ve hurt Pattie.” In fact, Mrs. Cogdon had, in her somnambulistic state, left her bed, fetched an ax from the wood heap, entered Pat’s room, and struck her two accurate forceful blows on the head with the blade of the ax, thus killing her. Mrs. Cogdon’s story was supported by the evidence of her physician, a psychiatrist, and a psychologist. The burden of the evidence of all three, which was not contested by the prosecution, was that Mrs. Cogdon had a form of hysteria with an overlay of depression and that she was of a personality in which such dissociated states as fugues, amnesias, and somnambulistic acts were to be expected. They agreed that she was not psychotic. They hazarded no statement as to her motives, the idea of defense of the daughter being transparently insufficient. However, the psychologist and the psychiatrist concurred in hinting that the emotional motivation lay in an acute conflict situation in Mrs. Cogdon’s relations with her parents; that during marital life she suffered very great sexual frustration; and that she overcompensated for her own frustration by overprotection of her daughter. Her exaggerated solicitude for her daughter was a conscious expression of her subconscious emotional hostility toward her, and the dream ghosts, spiders, and soldiers were projections of that aggression. The jury believed Mrs. Cogdon’s story and regarded the natural consequences of her acts as being completely rebutted by her account of her mental state at the time of the killing and by the unanimous support given to it by the medical and psychological evidence. She was acquitted. It must be stressed that insanity was not pleaded as a defense—Mrs. Cogdon was acquitted because the act of killing itself was not, in law, regarded as her act at all.

As stated in the case documents, this case “illustrates the impossibility of…satisfactorily…sever[ing] ‘act’ from ‘intention’.…Thus, Mrs. Cogdon’s action not being ‘voluntary,’ no question of criminal liability arose.” Mrs. Cogdon escapes basically because of the state of her consciousness,


75

not because she had no conscious intention or rational motive to kill, a state she shares with many convicted murderers. She was “asleep”; had she been “awake” her only defense would have been one of insanity. However, the difference between being asleep and being awake is not absolute. Consciousness is not like a light, either off or on; it is a finely graded scale ranging from death to the extreme awareness of the artist. Only instruments such as an electroencephalograph can chart certain variations of consciousness between people and in one person at different times. Had Mrs. Cogdon been awake—that is, just a little more conscious, a little more aware of her actions—then her act may have had to be regarded as voluntary. The line is an extremely fine one, as is shown by the fact that in and during her dream Mrs. Cogdon was “aware” of the ax, her daughter, and the soldiers. Not unexpectedly, she could not remember this part of the dream after the event, for often people struggle to repress profoundly disturbing and shocking memory traces. Thus all people dream, but some, for various reasons, remember more than others. Nor would Mrs. Cogdon’s position have been legally different if she could have recalled all of the dream, including the killing. Her exculpation lay not in the state of her memory but in her inability to bring into consciousness her emotional motivations and, consequently, her diminished awareness of the deed.

Hormonal Theories Another group of cases raising issues of altered biology in connection with a homicide are those related to hormonally altered states at the time of the homicidal act.

Premenstrual Syndrome One recent area of disagreement has centered on the diagnosis of premenstrual syndrome (PMS) and its relationship to violence. This is not an official diagnosis in DSM-IV-TR,34 and it is listed in an appendix of categories for which insufficient information exists for inclusion as an official diagnosis. Hence DSM-IV-TR lists PMS as “research criteria for premenstrual dysphoric disorder.” The argument rages as heavily as the hormones themselves are alleged to rage. Those who argue affirmatively hold that PMS is a hormone deficiency state that renders perpetrators unable to control their actions when these symptoms are present. Some now describe PMS as a mental disorder, but this description is contrary to its use as a legal defense.35 The medical profession continues to have diverse views about whether PMS is a syndrome or an illness.

76


One analogy drawn is between PMS and the phenomenon described over the centuries as postpartum psychosis and infanticide. Presentations of postpartum psychosis can vary from mild depression to outright delusional states with hallucinations. The occurrence of postpartum symptoms can vary from a few days to months after parturition, which suggests an etiology based on hormonal alterations, although psychosocial stressors cannot be ignored. However, the argument does not stop with the suggestion of hormonal alterations. It is alleged that the behavior from such a physiological disturbance does not fit the pattern of a mental illness, and the acts are more in the nature of automatist acts related to altered states of consciousness. PMS merits further discussion in view of its increasing publicity and in anticipation of its increased future use in legal settings. Medical testimony, as well as critiques by social commentators, will be the deciding factor in its use or abandonment until more definitive court rulings are obtained. Some feminist critiques have raised questions regarding the disease status of PMS. The objection to seeing PMS as a specific hormonal imbalance that can be medically managed is that the explanation is overly simplistic and premature. Such an approach views the problem in terms of a body needing repair rather than someone immersed in a genderbased social system that induces problems. Others challenge that PMS is not a valid clinical entity and that the causal explanations proffered to account for its syndrome status are insufficient. Those arguing in favor of it being viewed as a disorder are seen as biased in their observations and explanations leading toward a linear, reductionist type of thinking. Questions have also been raised about reports cited in support of establishing a premenstrual syndrome. Many of these reports lacked control subjects; some research suggested no difference between hormonal profiles of females alleged to have PMS and those who do not. Other investigators have confused statistical variation connected with premenstrual changes with the presence of an abnormal state. Different studies have produced a seemingly endless list of symptoms compatible with almost any diagnosis, or with no diagnosis in particular. One study listed 150 symptoms categorized into five groups: affective, neurovegetative, central nervous system, cognitive, and behavioral disturbances.36 When any of these occur in the premenstrual period, the supposition is that the symptoms are attributable to the syndrome. In 1961, Dalton37 studied 156 English women who were incarcerated and who had committed a crime in the previous 28 days. She divided the menstrual cycle into seven 4-day segments and found that 49% of the women were either in the premenstrual phase (4 days before menstruation) or menstruating (the next 4 days) at the time of their crimes. Her


77

conclusion was that hormonal changes (e.g., increased aggression, irritability, labile emotions, lethargy, carelessness) either caused the actions or affected the women’s ability to refrain from crime. However, Horney38 challenged this interpretation by arguing that committing the criminal acts, and the stress of arrest and imprisonment, could trigger early menstruation. In addition, she pointed out that 155 of the 156 women had been convicted of nonviolent crimes. The safe conclusion is that the symptoms of PMS as an explanation for violent crime are more correlated with publicity than empirical data.

Other Considerations Regarding Hormonal Factors There is merit in investigating the possibility that some type of hormonal changes occur in relation to an individual’s proneness to violence. However, several impediments also exist in documenting relevant hormonal levels and their effect on an individual’s actions. Investigations into the connection between hormones and violence often attempt to determine whether higher levels of certain hormones are present in those who exhibit violent behavior. However, the simplicity of such research may conceal many of the medicolegal problems that can arise from too-easy causal attributions. For example, establishing a murderer’s hormonal levels at random intervals does not indicate whether the level may actually have been abnormal at the time of the murder. In addition, simply obtaining a hormonal level at any one time does not reflect the totality of the individual’s endocrine functioning; the complex interactions within the hypothalamicpituitary-gonadal system must also be considered, such as concentrations of hormones in the brain, low levels of neurotransmitters, different sensitivities at receptor sites, and postreceptor effects. The confounding influence of alcohol and nonmedical drugs also can alter results of hormonal assays and blood levels of certain sex hormones. Another possibility is the connection between hormonal levels and diverse endocrine diseases (e.g., diabetes with hypoglycemic episodes, Cushing’s syndrome, hyperthyroidism). Results from a Swedish study of “normal” 16-year-old adolescents in which the subjects reported on their verbal and physical aggression in response to what they saw as provocations indicated a significant association between the aggressive behaviors and plasma testosterone levels.39 A few years earlier, investigators had found elevated levels of plasma testosterone in 52 rapists and child molesters.40 The level of testosterone was correlated with the degree of violence and the symptoms self-reported on an inventory. In a later study, some of those same investigators found no

78


significant differences in testosterone levels between a control group and a violent group.41 Again, a major deficit in such studies is that the level of hormones at the time of the violent act is not available. Although it may be deemed impossible that such levels could be accurately attained, the inference is nevertheless made by advocates that the hormonal level could have been abnormal at the time of the violence. Critics of any hormone–violence causal relation argue that 1) diverse causes are possible for violence even when altered plasma levels of hormones are found and 2) no causal connection can be drawn between elevated hormonal levels and violence. Perhaps these arguments suggest why authors of many improvised studies using subjects from prison populations have failed to find significant correlations. At a minimum, a baseline level of androgens would have to be established for the participants in these studies. Of those studies that have focused on hormonal levels and violence, at least three have found no connection. One group of investigators found that when a synthetic form of progesterone or estrogen was administered to sex offenders, it had no noticeable effect on their aggressive behavior.42 Another group who studied thousands of hostile patients in state hospitals concluded that there was no difference between men and women in such studies.43 Similarly, researchers in a later study failed to find any differences in androgen levels, particularly testosterone, among murderers, assaulters, and a control group.44 Green,45 in the book Sexual Science and the Law, thoroughly reviewed the simplistic thinking that leads to attempts to control serious sexual violence by way of either surgical or hormonal castration. Proponents of such intervention have argued that if the effects of androgen are blocked, sexual behavior should be blocked and presumably any connected violence. Unfortunately, the degree of aggression and testosterone levels are poorly correlated. Although results from some European studies on surgical castration have indicated less recidivism in castrated sex offenders compared with those who did not undergo such treatment, the results have been variable, especially if the castrations were carried out on young males. There would also be much legal and social objection to such procedures in the United States. As a result, studies in the United States have focused on treating sex offenders with medroxyprogesterone acetate (Depo-Provera) and outside the United States with cyproterone acetate (CPA). One difficulty in discussing this work with respect to homicidal violence is that the legal category of sex offenders is extremely heterogeneous, and only a small number of offenders have acted in homicidal ways. Hence, whether these drugs can reduce homicidal levels of violence in connection with sexual acts is unknown.


79

Critical analyses could similarly be presented regarding other hypotheses about hormonal possibilities, such as those involving the hypothalamic-pituitary axis, the thyroid, or the adrenal glands.

Posttraumatic Stress Disorder Consideration of automatisms possibly related to seizure disorders or altered hormonal states has now been extended to consideration of posttraumatic stress disorder (PTSD) as an exculpating (or at least mitigating) condition in homicidal behavior. The reasoning behind this concept is that some type of traumatic event occurs that alters the physiological, as well as the typical psychological, functioning of an individual (see discussion of abused persons in Chapter 7, “Masochism and Homicide”). A host of symptoms are then said to emerge, such as memory loss, sleep impairment, nightmares, reliving of the traumatic event, intrusive thoughts, exaggerated startle response, decreased emotional responsiveness, feelings of alienation, and dissociative conditions. The idea is that a peritraumatic dissociation may have occurred that elicits a variety of psychophysiological reactions. A dissociative state could be seen as a freezing response in contrast to a fight-or-flight response.46 In the dissociated state, the individual is portrayed as reliving the event as if he or she were once more present as a participant. Such reasoning was promoted originally for veterans of the Vietnam War who had been in combat, were self-medicating themselves with narcotics or alcohol, and subsequently committed violent assaults, including homicides. The automatism defense was often raised in this context using this diagnosis. Although the concept of a traumatic neurosis is not new, its use in this form as exculpatory for homicides is. Individuals involved in diverse types of traumas may have come from unfortunate social backgrounds. These traumas of developmental origin are used in arguing that a homicide occurred during an automatized state because the individual was predisposed to committing violence from childhood traumas. The possible traumas that can be used for legal defense purposes are endless and include past physical altercations that induced a traumatic state; being assaulted; fearing repeated assaults and thus having a need to strike out first; and being abused as a child in either a physical, sexual, or emotional manner. Use of the battered woman syndrome or the battered child syndrome as legal defenses to a homicide may raise the issue of PTSD. To complicate matters, perpetrating a homicide itself can lead to the onset of PTSD. In one study, there was a 58% lifetime prevalence of PTSD after a homicide.47

80


The idea of automatized homicides associated with earlier traumas may be extended to include homicides occurring subsequent to the emergence of multiple personalities or brainwashing events. However, assertions that an automatism defense is equivalent to that provided by “a disease of the mind” are questionable. This branches into quite a different pathway in terms of a legal defense, and for what follows if this defense is successful in court against homicide charges. There would be minimal justification, in the absence of any disease, to engage in a civil commitment process and hospitalize someone for a stress syndrome after such an acquittal.

Hypoglycemic States Similar complexities beset the assessment of hypotheses about a connection between hypoglycemic states and violent behavior. Studies of this relationship are often not comparable, with some referring to violent behavior, others to states of aggressiveness, and others to antisocial behavior. Some simply report the presence of irritability associated with hypoglycemic episodes. A distinction is also made between the sources of hypoglycemia, such as from physiological variations after an excessive release of insulin on a postprandial basis (also called functional hypoglycemia) in contrast to a state of fasting hypoglycemia, which may be associated with islet cell tumors or chronic alcohol abuse. The only safe conclusion at present about hypoglycemic states is that such states are only one of many other variables operating in aggressive behavior, including personality variables as well as the entire phenomenon of interpersonal reactions present in many social situations. Although a rare case of extreme violence may be connected with hypoglycemia, it is so entwined with other factors that it is difficult to achieve any causal specificity.

Role of Neurotransmitters Some current hypotheses have focused on serotonergic neurotransmitter processes involving aggression. Since Asberg and colleagues’48 1976 study in which it was found that patients with the lowest cerebrospinal fluid (CSF) 5-hydroxyindoleacetic acid (5-HIAA) concentrations (the serotonin [5-HT] metabolites) were most likely to attempt or commit suicide by violent means, there have been subsequent confirmations. Investigators in one study reported CSF 5-HIAA concentrations as being inversely related to the incidence of overtly aggressive behaviors.49 Rather than being used to focus on a specific diagnosis, the question of altered serotonin metabolism has entered the picture in attempts to ex-


81

plain violence. Two other groups assessed CSF 5-HIAA levels in men connected with a homicide and men who had attempted a suicide.50,51 Both groups found that lower levels of CSF 5-HIAA prevailed in those with a history of aggressive behavior as well as suicidal behavior. Maiuro and Eberle52 studied men who had killed or attempted to kill in a cruel manner; the authors first excluded those with diagnoses of schizophrenia or major affective disorders and found that the remainder had a lower level of CSF 5-HIAA. Brown and colleagues53 reported that young adults with low levels of CSF 5-HIAA tended to have a childhood history of aggressive conduct disorder that included acts such as lying, stealing, killing a pet animal in a rage, and impulsive fire setting. Exactly what implications these study results have for homicidal violence are unclear, although intriguing findings continue to be made involving altered serotonin metabolism, as well as altered levels of dopamine and noradrenaline. The basic idea is that something has gone awry with serotonin metabolism, whatever the attached diagnosis might be, and that the serotonergic aberration may represent a state or trait phenomenon. Other factors are seen as indirectly involving 5-HT neurotransmission. The theory is that the propensity for impulsive aggression is associated with activating negative affect. There is then a failure to respond appropriately to anticipated negative consequences of behaving aggressively.54 The key question is, what would make some individuals more vulnerable neurochemically? This neural circuit is composed of several regions (the prefrontal cortex, amygdala, hippocampus, medial preoptic area, hypothalamus, anterior cingulate cortex, insular cortex, ventral striatum) and interconnected regions. These multiple regions are dependent on the actions of neurotransmitters, such as 5-HT. The 5-HT system is seen as regulating aggression in animal models and violent behavior in humans.54 Hence in studies, low concentrations of the metabolite, 5-HIAA, are reflected in impulsivity and aggression. Diverse genetic influences, such as the 5-HT system, and added environmental stressors, such as the frequent influence of drugs or alcohol, are related as well.

Cerebral Hemisphere Dysfunctions The increase in research at the micro level looking for genetic aberrations related to neurochemical difficulties on the macro level is fairly recent; however, research on aberrant functioning in the cerebral hemispheres, in which anatomical and physiological factors relevant to violence have been studied, has had a much longer history. This latter type of research has waxed and waned throughout the 1900s.

82


Focusing on the cerebral hemispheres as the site of violent behavior seems to have had its modern origins by way of phrenology. The English forensic psychiatrist W. Norwood East56 pointed out in a 1928 lecture how the influence of Francis Joseph Gall (1758–1828) as a phrenologist emphasized the manifestations of the mind with the development of the brain. The brain was viewed as an aggregate of parts that had special functions. The idea was that the size of particular parts of the brain bore a relationship to the energy associated with particular mental powers. It was hypothesized that if certain “organs” of the brain were large, as deduced from the skull, particularly for such features as combativeness, destructiveness, and acquisitiveness, the organs and dispositions would correspond. This type of thinking led some leading physicians of the nineteenth century to apply these ideas to criminal behavior. East56 pointed out that Philippe Pinel had a patient in 1791 with recurring attacks of an overpowering impulse to assault the first person he saw. It was cited by the phrenologist George Combe as reflecting an “organ of destructiveness” being excited into diseased activity. Combe had compared the conduct of executed criminals with their cerebral development. The idea was that if “organs of the mind” acting through faculties of cerebral origins cannot adequately guide conduct, questions of responsibility for the behavior arise. These issues continued to be debated throughout the century, and by the late nineteenth century Cesare Lombroso (1836–1909) was looking for degeneracy of criminals in their physiognomy. Lombroso was a physician employed in the Italian penal system who founded the school of positivistic criminology. He focused on lateral asymmetries, such as abnormal asymmetries in the face and imbalances in the cerebral hemispheres.57 In fact, the positivist revolution has been extended from its origins in biology to the current state of biological criminology. The failure to confirm this type of theorizing as time went on led to its abandonment and ridicule. However, the idea that something could be impaired in the cerebral hemispheres of someone who committed a violent act such as murder was implanted, to be pursued later in current research and clinical studies. Parallel to these anatomical and physiological formulations about the cerebral hemispheres and violence, clinical findings were documenting changed behavior after cerebral injuries. The case of Phineas Gage in 1848 is often cited as beginning the inquiry. An explosion drove an iron tamping rod through his left cheek and out the top of his head, leaving a 1.5-inch hole in his frontal lobes.58 Previously a conforming, self-contained 25-year-old, Gage became fitful and unrepentant about his actions, showed no deference to others, and became impulsive and hedonistic. In 1994, Damasio and colleagues59 used computer-assisted


83

imaging to reconstruct what had actually happened to Gage, namely a disconnect to most of his frontal cortex that has been referred to as the first accidental lobotomy. Many subsequent clinical cases have been pursued since then, showing that frontal lobe dysfunctioning can be attributed to diverse etiologies, which vary from brain tumors to concussive types of brain injuries, frontal lobe epilepsy, and so on.60 These lesions may leave the individual prone to impulsive and aggressive acts, particularly if there has been damage to the anterior cingulate or orbital frontal cortex. Head injuries can contribute to several adverse consequences involving violence. Increased aggressiveness due to frontal lobe impairment, varieties of episodic dyscontrol as noted earlier, and exaggeration of preexisting antisocial behavior in individuals already vulnerable to antisocial behavior are a few possibilities that can lead to homicidal behavior. Lesions in the anterior part of the hypothalamus are noted for rage reactions, especially in those persons with a low threshold for being provoked. The existence of lesions may indicate a dysfunction in warning these individuals when to avoid certain provocative situations.61 Currently, researchers view biological abnormalities as perhaps one variable among many that can produce an individual vulnerable to violent acts. More recent hemispheric research has delineated differential features between the functioning of the left and right hemispheres, with the left operating in the area of sequential information processing on a verbal level, and the right hemisphere functioning on the level of mediating spatial stimuli. In the early 1970s, Flor-Henry62 questioned whether psychopathy was associated with left hemisphere disturbances. In a subsequent study, Yeudall63 found 72% of violent criminals had left hemisphere dysfunction compared with 79% of nonviolent criminals, who had a right hemisphere dysfunction. Yeudall’s study was subsequently criticized on the basis that the subjects studied were gathered from a criminal population, which would be expected to have a greater loading of individuals with neurological abnormalities, and that the criteria used to diagnose hemispheric dysfunction were not specifically spelled out.64 Along this same line, Nachshon and Denno65 investigated the hypothesis that “delinquent subjects” would show a higher incidence of left-sidedness compared with a control group when measured in terms of hand, eye, and foot preferences. Although delinquent subjects may be a long way from homicidal subjects, one of the six groups Nachshon and Denno studied was classified as very violent. This group showed a significantly higher incidence of left eye preference, interpreted as a left hemisphere dysfunction by the authors. These results did not demonstrate that such a hemispheric asymmetry exists in someone who mur-

84


ders; however, these authors’ results and other empirical studies are intriguing because of the paucity of much experimental data dealing with extreme human violence. Although it would be presumptuous to jump to a conclusion that a statistical association between hemispheric dysfunction and certain kinds of violence establishes any kind of causal relationship, certain questions about the meaning of the statistical association are relevant. One possibility is that a right hemisphere dysfunction could impinge on emotionality, which would contribute to impulse control problems. A left-sided hemisphere dysfunction could upset the role and use of language and coping skills in controlling violent impulses. Foresight and anticipation could also thereby be impaired, perhaps as seen in the prevalent diagnosis among male children with attention-deficit hyperactivity disorders. These types of deficits become more ominous as adolescence is approached, and the effects continue into adulthood in some people. The ongoing effect of these deficits may all lead back to questions about frontal lobe syndrome, the effects of head injury, early childhood head injuries, an individual’s proneness to overreact to alcohol, frontal lobe connections to the anterior-temporal cortex, impaired blood flow to parts of the medial and frontal cortex, and hippocampalamygdala areas connected to some type of central nervous system diathesis for violence. Deficits in these processes tend to alter the way an individual perceives the world and uses techniques to adapt and cope. In that sense, these deficits interfere with the effectiveness of psychological mechanisms and defenses and lead to more difficulty in exerting controls. In a primary preventive sense, determining what prenatal, perinatal, and postnatal factors may induce such deficits is crucial for this subgroup. In a secondary preventive sense, social and emotional assets can sometimes limit the effect of these disabilities. Here, traditional emphases have been used on ameliorating socioeconomic conditions, resolution of family conflicts, intervention with special schooling and counseling for those with deficits in cognitive functioning, and supportive therapies. However, the lack of consistent use of these interventions at a high caliber of performance, and over an extended duration of time, has limited their effectiveness in a majority of cases. A variation is the study of cerebral blood flow (CBF) via positron emission tomography in violent psychiatric patients compared with nonviolent groups. These studies show a significant decrease in CBF in the medial temporal and frontal cortex areas compared with a control group.66 Similar findings were made in 16 out of 21 aggressive individuals, indicating a hypoperfusion in the frontal or temporal lobes as


85

shown by single photon emission computed tomography.67 Positron emission tomography studies of CBF in a group of murderers who pleaded legal insanity were conducted by Raine and colleagues,68 and again a finding of lowered CBF in the prefrontal cortex was found in the study group compared with a control group. Reduced prefrontal cortex functioning was later shown to be connected to those showing affective aggression rather than instrumental aggression. Neuropsychological work has shown that only lesions in the orbital frontal and medial frontal cortex, rather than in the frontal cortex more generally, lead to disinhibition of the subcortical circuitry mediating affective aggression.69

CONCLUSION Flaws in most neuropsychological and neuropsychiatric research on violent behavior make generalizations precarious. Moffitt70 has done a thorough analysis of these limitations, which can only be highlighted here. In most research involving crime, the lack of an adequate control group is striking. Certainly this absence is true for studies of extreme degrees of violence, such as homicides. At the other end are clinical reports that at best offer excellent insights and hypotheses. In fact, perhaps congruent with positivistic criminology, the more microscopic the examination of an individual murderer, the more difficult it is to generalize beyond what may seem idiosyncratic. Although killing is still a relatively rare event, it does mean that there is often a conflict between seeking in-depth knowledge about why a particular individual behaved in that manner versus knowledge using larger numbers and listing diverse variables. In empirical studies, which often emphasize methodology and statistical analysis, those included in “violent groups” are almost always males, which parallels their prevalence in any study of crime. However, there is a comparative lack of data for females. A more serious flaw is that there are rarely adequate controls for demographic and environmental data. Some studies do not even give the age, gender, or race of those studied. The range of ages in studies may be enormous. In some communities, race has been eliminated from databases on the premise that collecting such data in itself is an example of racism rather than an attempt to isolate significant variables to remedy the situation. Although socioeconomic status may be given, the impact of this variable on prenatal or perinatal factors is often ignored. What is inescapable in studies of homicidal violence is that studying the perpetrators is always a retrospective event. Although it would be


86

possible to do some longitudinal prospective studies, these would focus on certain behaviors and traits related to aggression and violent behavior predicted to be related to homicide. Such studies have been done, but they would be next to impossible to perform for a group of potential homicide perpetrators. The limitations of data obtained from someone awaiting trial for a homicide or raising an insanity defense have long been known. Similar limitations exist for studying those in prisons or mental hospitals who have committed homicides. The result is that most of our studies are retrospective in nature. Where do these quandaries leave us? In one of the most complete models of studying future adult criminality, it was found that biological and environmental variables can predict only 25% of future adult criminality for males and 19% for females.71 I note that although statistical significance was attained in the study, the results still left 75%–81% of the criminal behaviors unexplained. Despite the statistical techniques in that study being as good or better than most social science studies, I question whether varying degrees of free will and determinism must be acknowledged with respect to criminal actions. These factors especially underscore that dichotomies between the biological and social factors are passé. Now the concept of a “biosocial” approach to homicidal violence becomes relevant. The ideal would be prospective longitudinal studies considering general biological and psychosocial influences on serious violence. This endeavor would entail testing specific hypotheses derivable from the theory, focusing on key variables to be studied, and detecting developmental sequences and causes.72

REFERENCES 1. 2. 3. 4. 5.

McDougall W: An Introduction to Social Psychology. New York, Barnes & Noble, 1960 Bibring E: The development and problems of the theory of the instincts. Int J Psychoanal 22:102–131, 1941 Lorenz KZ: On Aggression. New York, Harcourt, Brace & World, 1966 Tinbergen N: The Study of Instinct. Oxford, England, Clarendon Press, 1951 Hollander E, Posner N, Cherkasky S: Neuropsychiatric aspects of aggression and impulse control disorders, in The American Psychiatric Publishing Textbook of Neuropsychiatry and Clinical Neurosciences. Edited by Yudofsky SC, Hales RE. Washington DC, American Psychiatric Publishing, 2002, pp 579–596

Biological Factors in Homicide 6.

7. 8. 9.

10.

11. 12. 13. 14. 15. 16. 17.

18. 19.

20. 21.

22. 23.

24.

87

Kupfermann I: Genetic determinants of behavior and learning and memory, in Principles of Neural Science, 3rd Edition. Edited by Kandel ER, Schwartz JH, Jessel TM. New York, Elsevier, 1991, pp 987–1008 Bioulac B, Benezech M, Renaud B, et al: Biogenic amines in 47, XYY syndrome. Biol Psychiatry 15:917–923, 1980 Cadoret RJ, Yates WR, Troughton E, et al: Adoption study demonstrating two genetic pathways to drug abuse. Arch Gen Psychiatry 52:42–52, 1995 Pardes H, Kaufmann CA, Pincus HA, et al: Genetics and psychiatry: past discoveries, current dilemmas, and future directions. Am J Psychiatry 146:435–443, 1989 Brock DW, Buchanan AE: The genetics of behavior and concepts of free will and determinism, in Genetics and Criminality. Edited by Botkin JR, McMahon WM, Frances LP. Washington, DC, American Psychological Association, 1999, pp 67–81 Montana v Egelhoff, 518 US 37 (1996) People v Yukl, 372 NY2d 313, 319 NY Sup Ct (1975) Mark VH, Ervin FR: Violence and the Brain. New York, Harper & Row, 1970 Pincus JH, Tucker GJ: Behavioral Neurology, 3rd Edition. New York, Oxford University Press, 1985 Crichton M: The Terminal Man. New York, Knopf, 1972 Delgado-Escueta AV, Mattson RH, King L, et al: Special report: the nature of aggression during epileptic seizures. N Engl J Med 305:711–716, 1981 Hermann BP, Whitman S: Behavioral and personality correlates of epilepsy: a review, methodological critique, and conceptual model. Psychol Bull 95:451–497, 1984 Monroe RR: Behavioral Disorders. Cambridge, MA, Harvard University Press, 1970 Gregg TR: Cortical and limbic neural circuits mediating aggressive behavior, in Neurobiology of Aggression. Edited by Mattson MP. Totowa, NJ, Humana Press, 2003, pp 1–20 Weiger B, Bear D: An approach to the neurology of aggression. J Psychiatr Res 22:85–89, 1988 Kohn SR, Asnis GM: Aggression in psychiatric disorders, in Neurobiology of Aggression. Edited by Mattson MP. Totowa, NJ, Humana Press, 2003, pp 135–149 Whitman S, Hermann B, Black R, et al: Psychopathology and seizure type in children with epilepsy. Psychol Med 12:843–853, 1982 Achenbach T, Edelbrock C: Manual for the Child Behavior Checklist and Revised Child Behavior Profile. Burlington, VT, Thomas M Achenbach, 1983 Lewis DO, Pincus JH, Shanok SS, et al: Psychomotor epilepsy and violence in a group of incarcerated adolescent boys. Am J Psychiatry 139:882–887, 1982

88


25. Hermann BP, Whitman S: Behavioral and personality correlates of epilepsy: a review, methodological critique and conceptual model. Psychol Bull 95:451–497, 1984 26. Mendez MF, Doss RC, Taylor JL: Interictal violence in epilepsy. J Nerv Ment Dis 181:566–569, 1993 27. Raine A: The Psychopathology of Crime: Criminal Behavior as a Clinical Disorder. New York, Academic Press, 1993 28. Tonkonogy JM: Violence and temporal lesion: head CT and MRI data. J Neuropsychiatry 3:189–196, 1991 29. Gunn J: Medical-legal aspects of epilepsy, in Epilepsy and Psychiatry. Edited by Reynolds EH, Trimble MR. Edinburgh, UK, Churchill Livingstone, 1981, pp 165–174 30. People v Grant, 46 Ill App 125, 360 NE 2d 809 (1977) 31. Model Penal Code, Section 2.01, American Law Institute, Philadelphia, PA, 1962 32. Eigen JP: Unconscious Crime. Baltimore, MD, Johns Hopkins University Press, 2003 33. Morris N: Somnambulistic homicide: ghosts, spiders, and North Koreans. Res Judicate 5:29–32, 1951 34. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 35. Severino SK, Moline MJ: Premenstrual Syndrome: A Clinician’s Guide (Diagnoses and Treatment of Mental Disorders). New York, Guilford, 1989 36. Moos RH: A typology of menstrual cycle symptoms. Am J Obstet Gynecol 103:390–402, 1969 37. Dalton K: Menstruation and crime. Br Med J 2:1752–1753, 1961 38. Horney J: Menstrual cycles and criminal responsibility. Law Hum Behav 2:25–36, 1978 39. Olweus D, Mattsson A, Schalling D, et al: Testosterone, aggression, physical, and personality dimensions in normal adolescent males. Psychosom Med 42:253–269, 1980 40. Rada RT, Laws DR, Kellner R, et al: Personal testosterone levels in the rapist. Psychosom Med 38:257–268, 1976 41. Rada RT, Kellner R, Stivastava C, et al: Plasma androgens in violent and nonviolent sex offenders. J Am Acad Psychiatry Law 11:149–158, 1983 42. Money J, Dalry J: Sexual disorders: hormonal and drug therapy, in Handbook of Sexology. Edited by Money J, Musaph H. Amsterdam, The Netherlands, Elsevier/North Holland, 1977, pp 1303–1310 43. Tardiff K, Sweillam A: Assault, suicide and mental illness. Arch Gen Psychiatry 37:164–169, 1980 44. Bain J, Langevin R, Dickey R, et al: Sex hormones in murderers and assaulters. Behav Sci Law 5:95–101, 1987 45. Green R: Sexual Science and the Law. Cambridge, MA, Harvard University Press, 1992


89

46. Brewin CR: Posttraumatic Stress Disorder: Malady or Myth? New Haven, CT, Yale University Press, 2003 47. Papanastassiou M, Waldron G, Boyle J , et al: Post-traumatic stress disorder in mentally ill perpetrators of homicide. Journal of Forensic Psychiatry and Psychology 15:66–75, 2004 48. Asberg M, Traskman L, Thoren P: 5-HIAA in the cerebrospinal fluid: a suicide predictor? Arch Gen Psychiatry 33:1193–1197, 1976 49. Brown GL, Goodwin FK, Ballenger JC, et al: Aggression in humans correlates with cerebrospinal fluid amine metabolites. Psychiatry Res 1:131–139, 1979 50. Lidberg L, Tuck JR, Asberg M, et al: Homicide, suicide and CSF 5-HIAA. Acta Psychiatr Scand 71:230–236, 1985 51. Linnoila M, Virkkunen M, Scheinin M, et al: Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Sci 33:2609–2614, 1983 52. Maiuro RD, Eberle JA: New developments in research on aggression: an international report. Violence Vict 4:3–15, 1989 53. Brown G, Kline W, Goyer P, et al: Relationship of childhood characteristics to cerebrospinal fluid 5-hydroxyindoleacetic acid in aggressive adults, in Proceedings of World Congress of Biological Psychiatry, Vol 7. Edited by Shagass C, Josiassen R, Bridger W, et al. New York, Elsevier, 1985, pp 177–179 54. Davidson RJ, Putnam KM, Larson CL: Dysfunction in the neural circuitry of emotion regulation: a possible prelude to violence. Science 289:591–594, 2000 55. Lesch KP: The serotonergic dimension of aggression and violence, in Neurobiology of Aggression. Edited by Mattson MP. Totowa, NJ, Humana Press, 2003, pp 33–63 56. East WN: The Relation of the Skull and Brain to Crime. Edinburgh, UK, Oliver and Boyd, 1928 57. Gottfredson MR, Hirschi T: A General Theory of Crime. Stanford, CA, Stanford University Press, 1990 58. Linno LA, Charney DS: The neurobiology of aggression, in Neurobiology of Mental Illness. Edited by Charney DS, Nestler EJ, Bunney BS. New York, Oxford University Press, 1999, pp 855–871 59. Damasio H, Grabowski T, Frank R, et al: The return of Phineas Gage: the skull of a famous patient yields clues about the brain. Science 264:1102– 1105, 1994 60. Stuss DT, Knight RT (eds): Principles of Frontal Lobe Function. New York, Oxford University Press, 2002 61. Niehoff D: The Biology of Violence. New York, Free Press, 1999 62. Flor-Henry P: On certain aspects of the localization of the central systems regulating and determining emotion. Biol Psychiatry 14:677–698, 1979 63. Yeudall LT: A neuropsychological perspective of persistent juvenile delinquency and criminal behavior: a discussion, in Forensic Psychology and Psychiatry. Edited by Wright F, Bahn C, Reiber RW. New York, Annals of the New York Academy of Sciences, 1980, pp 349–355

90


64. Hare RD: Psychopathy and laterality of cerebral function. J Abnorm Psychol 88:605–610, 1979 65. Nachshon I, Denno D: Violent behavior and cerebral hemisphere function, in The Causes of Crime: New Biological Approaches. Edited by Mednick SA, Moffitt TE, Stack SA. New York, Cambridge University Press, 1987, pp 185–217 66. Volkow ND, Tancreedi LR, Grant C, et al: Brain glucose metabolism in violent psychiatric patients: a preliminary study. Psychiatry Res 61:243–253, 1995 67. Soderstrom H, Tallberg M, Wikkelso C, et al: Reduced regional cerebral blood flow in nonpsychotic violent offenders. Psychiatry Res 98:29–41, 2000 68. Raine A, Buchsbaum MS, LaCasse L: Brain abnormalities in murderers indicated by positron emission tomography. Biol Psychiatry 42:495–508, 1997 69. James R, Blair R, Charney DS: Emotion regulation: an affective neuroscience approach, in Neurobiology of Aggression. Edited by Mattson MP. Totowa, NJ, Humana Press, 2003, pp 21–32 70. Moffitt TE: The neuropsychology of delinquency: a critical review of theory and research, in Crime and Justice: An Annual Review of Research, Vol 12. Edited by Morris N, Tonry M. Chicago, IL, University of Chicago Press, 1990, pp 99–120 71. Denno DW: Biology and Violence: From Birth to Adulthood. New York, Cambridge University Press, 1990 72. Farrington DP: Key issues in studying the biosocial bases of violence, in Biosocial Bases of Violence. Edited by Raine A, Brennan PA, Farrington DP, et al. New York, Plenum, 1997, pp 293–304

CHAPTER

3 SCHIZOPHRENIA, DELUSIONAL DISORDERS, AND THE PREDICTION PROBLEM REGARDING HOMICIDE

THE PROBLEM OF DETERMINING the relationship between homicide and schizophrenia is a subset of the broader problem of the relationship between violence and any type of mental illness. Making valid and reliable predictive statements about such a relationship is an elusive process. Some homicides are bizarre and can only be understood in the context of psychotic thought processes. Schizophrenic individuals or individuals with a blatant delusional system are often thought of as being “crazy,” with the media sometimes helping to foster this image. The connotation is that these people are running loose in the community and likely to be violent. One persistent scientific question, then, is, why does it remain so difficult to predict homicidal violence within this group of individuals who may be so disturbed? It is understandable that it is difficult to predict violence occurring in connection with other types of mental disorders, because individuals with those disorders do not act so strikingly bizarre, but it would seem that it should be easier to make predictions at least in connection with individuals who are exhibiting psychotic symptoms as part of their disorder. However, it is a very difficult process, and even more difficult regarding individuals who live quietly with their psychotic thoughts prior to a homicidal act.

91

92


PUBLIC IMAGE OF MENTALLY ILL INDIVIDUALS AS BEING “CRAZY” Males and, to an increasing extent, females who have engaged in various types of killings for which the motive has never been clarified are often simply placed in an amorphous category of “crazy killers.” This popular image of mentally ill people as being violent is shaped by the endless stream of novels involving violence and killing as well as television portrayals that are constructed from or based on actual killings. It is easy to see how this image was reinforced in the twentieth century, at least up through the 1950s, when mentally ill patients were involuntarily committed to psychiatric hospitals, and the public saw them as a group of strange people who needed to be isolated from the rest of the community. However, a study in the late 1990s in the American population found that 61% believed that people with schizophrenia are either “very likely” or “somewhat likely” to do something violent to others.1 One might have thought that by the twenty-first century, as mentally ill patients were gradually moved into communities and being maintained there with psychotropic medications, the public’s increased contact with them might have ameliorated their image problem. In reality, the assumption behind deinstitutionalization—that these patients would be treated effectively in the community setting—has largely remained unfulfilled, with the result that large numbers of homeless, victimized, mentally ill individuals live on the streets of all major American cities and cannot access needed treatment. Thus deinstitutionalization may have had the opposite effect on the public; rather than revising their image of mentally ill patients, members of the public, who lack the tools for understanding the plight of these people, have become more frightened from their increased encounters with the mentally ill in their communities. It should be noted that the image of mentally deranged individuals or “lunatics” being violent evolved in earlier centuries—a good deal earlier than the institution of involuntary hospitalization processes. Hence, it is not likely that the image problem will abate in the near future. The question remains as to whether mentally ill individuals merely carry the image of being crazy people and being more homicidally violent or actually are violent. Depending on the answer to this question, several specific implications are presented. One is whether the social policy and legal decisions that have shifted large numbers of mentally ill people into the community have been correct. If it is established that mentally ill individuals do in fact commit an inordinate number of homicidal acts, then community policies, practices for releasing mentally ill individuals into the community, and current civil commitment criteria will need to be reexamined.

Schizophrenia, Delusional Disorders, and Homicide

93

An interesting paradox arises when a comparison is made between mentally ill individuals who have committed violent acts and nonmentally ill individuals who have committed sexual offenses. Courts have upheld statutes and legislation related to psychopathic individuals who commit sexual crimes that allow them to be detained long after their criminal sentences have expired.2 Their extended confinement is justified on the pretext of a civil commitment process based on their alleged dangerousness, even though they have never been diagnosed with a major mental illness and treatment results for this legal category are questionable. Similar questions may arise about how schizophrenic individuals who commit homicide should be handled by the criminal justice system, even for those who reach a state of remission and then commit homicide. It may be argued that schizophrenic individuals who commit homicide deserve sentences as long as those currently being imposed on other, nonmentally ill individuals convicted of homicide in the United States. This would mean that in the case of first-degree murder, the sentencing under capital punishment should be carried out as provided in the statutes now on the books in 38 states. The public’s desire for retribution, accompanied by assumptions of the schizophrenic individuals’ potential and continued dangerousness, can be a powerful influence on juries and judges deciding punishments in terms of a justdeserts model. Studies of the prevalence of mental illness among homeless individuals have provided widely varying estimates, in part because the definitions of mental illness used are unclear. A further difficulty is that some studies on the prevalence of schizophrenia often exclude “street people” because they are more difficult to reach and include for survey purposes. Nonetheless, the studies that have been conducted seem to indicate that 30%–40% of homeless people may have a major type of mental illness. 3 When bizarre homicidal acts occur in communities, glaring headlines and coverage may focus on this group, with the specious rationalization by civil libertarians that such acts reflect or should reflect community norms of “the right to be crazy.” The New York case of Juan Gonzalez illustrates the problem. Gonzalez had initially told the staff at a Manhattan shelter, “I am going to kill. God told me so. Jesus wants me to kill.” After being sent to a hospital, he was released within 48 hours, and 48 additional hours he later stabbed nine people aboard the Staten Island Ferry, resulting in two deaths. Another case that garnered publicity was that of Jorge Delgado, who clubbed an usher to death at St. Patrick’s Cathedral in New York City after running through the church naked.4 Similar cases, with their dramatization of bizarre homicidal violence, continue to occur in every major American city.

94


VALIDITY OF PREDICTING VIOLENCE IN MENTALLY ILL INDIVIDUALS Overextension of Psychiatric Expertise The increasing interdependence of the legal and the mental health systems over time has become a double-edged sword. Psychiatric expertise has been expanded into areas beyond civil commitment and “not guilty by reason of insanity” issues, such as in the areas of diverse competency questions, assessment of sex offenders’ responsibility for their acts, assessment of sex offenders for treatment programs, or matters of sentencing and release from penal institutions or psychiatric hospitals. Psychiatrists, and increasingly other mental health workers, are asked to assess individuals’ dangerousness as well as diagnose any mental illnesses. The consequences of psychiatrists and others being asked to do more than they were obliged to do or might be capable of, in what many consider to be an overextension of their expertise, have been inevitable. For example, psychiatrists have begun to be held legally accountable for missing predictions of homicidal violence in civil cases. Perhaps this turn of events was to be expected, given the legal cases in which psychiatrists (and other mental health personnel in similar roles) have been found to have a duty to warn third parties of possible violence from patients. This legal burden was subsequently expanded to the principle of the “duty to protect” others, in which a psychiatrist was held to a standard of making a prediction of violence so that third parties would not be injured. These types of predictions would then be seen as mandatory exceptions to confidentiality analogous to laws that were enacted requiring physicians, or perhaps mental health professionals, to report certain communicable diseases, gunshot wounds, child abuse cases, or in some states narcotic addiction. The difference is that in cases involving violence potential, the professional may be held liable for not making a prediction. The academic problem of prediction must be distinguished from the practical one. Whether psychiatrists, or anyone else, can theoretically make valid predictions about the future occurrence of homicidal violence should be sharply distinguished from the practical situation of not making such a prediction and then subsequently being held legally liable for violating a duty to warn or to protect the victim.5 The denial of having such predictive talents goes to the issue of attempting to deny legal liability in malpractice cases for erroneous predictions or failure to make such predictions. The latter situation was amply discussed subsequent to the initial California case involving Tatiana Tarasoff, who was murdered by her ex-boyfriend several months after she broke off their


95

relationship.6 The mentally ill boyfriend had revealed his murderous thoughts to a psychologist at a student health service; the psychologist then notified the campus police, urging them to hospitalize the man. The police felt they had no basis to hospitalize him, but it was later held that the psychologist and supervising psychiatrist could be considered liable for not warning Tarasoff that her life was potentially in danger. The reasoning, after the fact, was that the clinicians should have been able to predict the homicide, at least to the extent that they should have warned the people who were enmeshed in the patient’s delusional and violent thoughts. The validity of making such predictions has been challenged on both legal and clinical grounds. The initial legal attack was a constitutional one, involving the use of the standard of dangerousness in the civil commitment of mentally ill individuals. The validity of a dangerousness standard of prediction was challenged both on substantive and due process grounds; in the latter arena, it was argued that the process of making such predictions could lead to an unreasonable deprivation of liberty. One by-product of the legal misuse of the prediction of dangerousness standard as the criterion used for civil commitment by courts and legislative enactments from the 1970s onward was the increasing tendency to turn psychiatric hospitals into facilities whose primary purpose was the detention of those who were labeled mentally ill and dangerous rather than the treatment of mentally ill individuals. The result of this change was clearly described by Stone,7 who referred to those “incarcerated” in the hospital system as a modal psychiatric hospital population composed of the mentally ill who were thought to be dangerous. Thus the hospital populations began to consist primarily of males ages 20–40 years who floated among different psychiatric hospitals, prisons, and the community. Many of these individuals presented the problem of having comorbid diagnoses and serious character disorders mixed in with episodes of psychosis. Substance abuse and dependence were also often present. Individuals with these combined psychiatric and social disturbances have an increased potential of being violent as well as being more resistant to neuroleptic medication.

Ability to Achieve Only Weak Predictability However, another initial theoretical denouement regarding clinicians’ predictions of dangerousness came in an article that used calculations to demonstrate that dangerousness was routinely overpredicted.8 A basic problem in dealing with predicting homicidal violence is that such

96


violence is a rare event, with a low base rate (i.e., a low incidence in the population). Because homicide itself is a rare event, a greater margin of error accompanies its prediction. Consider a hypothetical case in which the incidence of killing in a community is 1:1,000. Imagine that a test could predict those incidences with 95% accuracy. (Of course, no predictor test exists that is even close to making such accurate predictions, including those based on the best of any available clinicians’ assessments or even actuarial tables.) In this scenario, 100 of every 100,000 people would be killed, and the test would be able to predict 95 of those 100. However, another group of 4,995 people would be identified as potential killers who would not actually kill. To make sure no killer was on the loose, even with the 95% accurate instrument, the 95 plus 4,995 individuals would have to be segregated from the community. This means that 54 harmless people would be detained for every 1 person who was, in fact, dangerous, in order to ensure the community’s safety. Most individuals and groups would see this scenario as unacceptable by legal and social standards. By 1984, a “second generation” of predictors began to emerge who were less zealous in their predictions, recognizing the limits of accurate prediction.9 It had been hoped initially, perhaps too optimistically, that violence prediction could be examined from a relative rather than an absolute moral-political position, in which decisions to incapacitate people are not necessarily being made solely on clinical grounds. Predictions of violence are customarily one of two types: clinical appraisals or statistical-actuarial methods. Clinical appraisals rely on the experience and professional expertise of decision makers. Statistical or actuarial appraisals rely on weighting key characteristics that are specified in advance, then using those scores to make a statistical prediction. Statistical techniques generate rules based on analyses of data from prior cases to measure the accuracy of such rules for those prior cases as well as future cases. To the surprise of many clinicians, the actuarial method comes out ahead as a predictive technique, even over the wisest of clinicians’ judgments. One group of investigators reviewed nearly 100 comparative studies in the social science literature on the relative accuracy of clinical and statistical predictions of violence.10 In every study, the actuarial method equaled or surpassed the clinical method, sometimes by a substantial margin. Those clinicians with some grounding in probability theory try to claim that such group statistics cannot apply to single individuals or events. Yet to say that an individual has unique features not factored into the statistical prediction does not negate the possibility that he or she also has common features that feed into the generation of rules with predictive power.


97

The implication of this discussion is that if a bet were to be made about the best way to make predictions, the money should be placed on a statistical-actuarial approach. Such an approach would appear to have the greatest potential for a study of homicide prediction. Of course, even here the limitations on trying to predict that a schizophrenic individual will commit a homicide in the future overwhelmingly weigh against the predictions’ accuracy. More recently, the trend has been to try to make much more limited types of predictions, such as whether someone will be violent while in a psychiatric hospital unit, or what the relationship will be between a patient’s admission symptoms and his or her future assaultive behavior while an inpatient.11 Even a recent attempt to achieve the narrow objective of determining which patients brought to an emergency department could have been predicted as being dangerous concluded only that certain state variables were associated with patient dangerousness, such as the immediate expression of hostility, agitation, impulsivity, hallucinations, or mania.12 Demographic variables (e.g., age, gender, employment status) were not found to be related to patient violence when assessed in clinical settings. Note that these types of narrowly defined studies dealt with individuals already viewed as raising some concern about violence by virtue of their being brought to the emergency department of a hospital. This approach is a long way from making predictions in an open society about a future homicide. The scope of these past studies indicates how great a disparity exists between current scientific capability to make predictions and the expectations of predicting violence that have evolved within the legal system and the accountability that has been attached to those predictions. Given the fallibility of making predictions about dangerousness in general, as well as more specifically in psychotic individuals, the focus has now shifted to “assessments of risk.”

EARLY STUDIES OF VIOLENCE AND MENTAL ILLNESS Definitional Problems Articles addressing the relationship between mental illness and violence continue to flow. The original studies abounded in definitional and methodological problems. The term violence itself has been subjected to shades of interpretation, and many times a distinction between types and degrees of violence was not made beyond references to bodily contact between people. Some studies that focused on intrafamilial violence included slapping—of a spouse or a child—as an act of violence. Although slapping can be viewed as an act of violence at times, with such a definition it is not surprising that high figures are re-

98


corded. One example is the figure of 93% of American parents spanking their children to discipline them.13 The use of physical force is believed to be widespread in American families, involving not only children but also intimates, spouses, the disabled, and elders.14 At the opposite end, few of the studies have confined themselves to a homicidal level of violence, even with only one diagnosis in mind, such as schizophrenia. This raises a problem of having to extrapolate from various degrees of assaultive behaviors to homicides. There are also problems in securing agreement on the definition and criteria used for mental illness. With respect to schizophrenia, past studies have used different criteria than current research. Unresolved questions remain about what should be regarded as an illness (or disease) and what should be regarded as a disorder. Should studies be confined to acute psychotic episodes for schizophrenic individuals, or should they include a broader definition of anyone who was at one time diagnosed with schizophrenia? There are also the pervasive problems of comorbid disorders, such as the “intrusion” of diverse personality disorders or drug abuse problems into schizophrenia, and determining which diagnosis, or whether several diagnoses, operated with respect to a homicide event. Unfortunately, for purposes of seeking explanations, a linear relationship has often been assumed in many studies, something that rarely exists in reality. Causal attribution is a much more complex state of affairs, but a much more realistic assumption.

Hospital Studies The unreliability of many studies can be illustrated in early work that addressed the problem of mental illness and violence up to the midtwentieth century. Authors of those studies usually concluded that mentally ill individuals were less likely to be violent than the general population. However, all of the studies were carried out on hospitalized mental patients, with their previous arrest rates or court conviction records being compared with those of the general population. Because many hospitalized patients during the first half of the century were confined for long periods of time, this placed them in a special category of long institutionalizations, and few valid conclusions could hold up about their violent propensities. The years spent hospitalized often coincided with the years in which the patients’ violent behavior would have been at its peak if the patients had been at liberty in a community. During the mid-twentieth century, and before the 1955 era when tranquilizer medications were introduced, the presumption could be made that a certain level of violence occurred, even while schizophrenic


99

individuals were hospitalized, among patients or with staff. In contemporary hospital settings, follow-up comparisons could be made between one inpatient unit that used antipsychotic medications and another that did not, or some individuals who did receive medications and those who refused. Again, this approach would still be investigating a narrow group of hospitalized patients, relying on diverse criteria for schizophrenia, and then extrapolating those violent tendencies to “homicidal violence.” All of this is considerably removed from a connection between a specific mental illness, such as schizophrenia, and homicides occurring in communities, without even considering the complication of a comorbid illness.

Studies of Nonhospitalized Patients and Follow-Up Approaches After early efforts up to the mid-twentieth century to study hospitalized mentally ill patients and their tendency to act violently, different approaches were tried. One approach was to study a broad range of psychiatric patients, not just those who were hospitalized. Authors of these studies then sought to obtain data to determine the subsequent incidence or prevalence of violent behavior among those patients. The patients could have been once hospitalized but discharged at the time of the study, or they could have been patients living in the community. Many investigators tended to use arrest rates as the defining measure indicative of violence. This approach led to underestimates of violent incidents. A second approach focused on mentally ill individuals who had committed violent criminal acts and attempted to ascertain what psychiatric symptoms had been present at the time of the occurrence. This approach led to questions about what diagnosis was present because it became relevant to legal questions, such as whether an insanity defense should be raised, sentencing, and so on. A selective look at the approach investigating subsequent criminal acts of those labeled as psychiatric patients illustrates how early studies led to questionable conclusions that violence or criminal acts were less frequent among mental patients. Because the focus in this approach was on hospitalized patients, a majority of the “nonorganic” cases would have had schizophrenia diagnoses at that time. Authors of a 1922 study followed patients for a period of 10 years after their discharge. 15 In many ways, this was a marvelous piece of research for the time, being done before grants or subsidies were available for such research. Only 12 (1.2%) of the former patients were arrested during the follow-up period, which then led the researchers to an unwarranted conclusion about the low incidence of violence among mentally ill individuals.

100


Subsequent studies continued to focus primarily on arrest rates, also with the conclusion that because arrest rates were lower after discharge, violence was obviously lower in that population. The methodology of these studies had not changed by 1945, when investigators in one study concluded that being hospitalized contributed to lowered subsequent arrest rates.16 The sample population of 1,676 in that study noted 18.4% of subjects with a police record before hospitalization and 5.2% of subjects with a police record after release, with no homicides recorded. Investigators in another study followed 1,638 former psychiatric inpatients for 6–12 years subsequent to their discharge from a psychiatric unit. By 1958, at the conclusion of the study, none of those patients had been reported as having committed a homicide or having died by homicide.17 However, intervening variables were not considered in leading to such results. Thus, by 1970, these earlier studies were being criticized on the basis of their selective relevance, for not taking into account the decrease in criminal behavior that can occur with age, and for the lack of any control groups.18 During the 1960s, a shift in thinking had begun in which former hospitalized patients were thought to be just as likely to engage in violent behavior as those in the general population. With respect to certain crimes, findings began to emerge that these patient groups might even have a higher incidence of violence. Rappeport and Lassen19 collected data from all but one state hospital in Maryland on patients who were discharged in 1947 and 1957. The authors compared the discharged patients’ arrest rates with those of the general population based on data from the Federal Bureau of Investigation’s Uniform Crime Reports. Although the authors found no indication that former male patients had higher rates of murder or negligent manslaughter than did males in the general population, the former patients did have higher arrest rates for robbery than did the general population. Former female patients were found to have higher arrest rates for aggravated assault than females in the general population, but the former female patients had lower rates for murder and robbery.20 At that time, studies were also beginning to become more focused, such as addressing the question of the specificity of diagnoses. However, the contaminating influence of diagnoses comorbid with schizophrenia was still not given primary recognition. In a follow-up study of 1,141 male military veterans, 95% of whom carried a schizophrenia diagnosis, Giovannoni and Gurel21 found that these veterans’ rate of violent behavior—as seen in homicide, aggravated assault, and robbery— exceeded that of the general population. Such studies began to draw attention to the problem of comorbidity, in that a subgroup of the schizo-


101

phrenic individuals was noted to have problems with alcohol. By 1978, Sosowsky,22 reporting arrest rates of 201 California state mental patients compared with the arrest rates of people in the patients’ local counties, found that those patients had a higher arrest rate for violent offenses, both before and following their hospitalization, than did the general population.

Retrospective Approach Another approach used to study the relationship between serious mental illness and violence has been to use a research strategy that begins with a designated group of individuals who has already committed a criminal offense. Some studied were under arrest or charged with a crime, whereas others had already been convicted of multifarious offenses. Again, only some of the offenses were homicides, so the investigators tended to group homicides and other assaultive behavior under the heading of “violence.” The methodology used in these studies has been varied, including checklists, psychological testing, and descriptive types of diagnoses. Occasionally, some phenomenological or psychodynamic reports have been used, but these were more in the nature of illustrating the subjects’ psychopathology rather than lending themselves to obtaining statistically significant results. An example of this approach was seen in the work of Guze and Cloninger, in which 209 convicted male felons were studied with a focus on their descriptive diagnoses.23,24 An interesting array of diagnoses was present, with sociopathy, alcoholism, and drug dependence being found more frequently in this group than in the general population. This finding confirmed the impression that forensic psychiatric professionals had long held about the presence of personality disorders and substance abuse in a criminal population. However, after 8–9 years of follow-up, Guze and Cloninger found that group members who had diagnoses of schizophrenia, manic-depressive disease, or organic brain damage did not have a higher incidence of violence than did the general population. A similar 2- to 3-year follow-up of female felons revealed they had higher rates of sociopathy, alcoholism, drug dependence, and hysteria. Yet the more serious mental illnesses, such as schizophrenia and manic-depressive illness, were not reported in these females. Although the approach of grouping felons together and then diagnostically assessing them is interesting, it does not provide information on whether certain combined diagnoses may have been the predisposing variable in homicidal behaviors.

102


The selectivity factor for homicide could be pursued by focusing on those arrested or convicted only for homicide, but this has rarely been done in any systematic manner. To implement such a project would require large numbers of individuals who would need to be assessed for multiple variables. A combined group would be composed of those arrested and perhaps taken to psychiatric hospitals or found incompetent to proceed to trial as well as those who were kept in the criminal justice system. Individuals who were arrested but not tried for homicide should not be conflated with those who were convicted or those who were found not guilty by reason of mental illness for a homicide. Once again, the problem of comorbidity looms as a pervasive problem, especially with the high rates of substance abuse and dependence present in such a cohort. A study involving the Bellevue catchment area in New York City found that in a group of mental patients, the presence of alcohol and drug dependency doubled those patients’ chances of being arrested.25 Again, the plaguing question remains as to whether a major mental disorder, such as schizophrenia, or a comorbid condition is making the major contribution to the propensity to violence—homicide or otherwise. Alternatively, the major contributor could be a factor other than a psychiatric diagnosis. A partial answer to this question came in Swanson and colleagues’26 study of violence and psychiatric disorder in the community using the database from the Epidemiologic Catchment Area (ECA) study. In the ECA study, based on household surveys, mental illness was seen as a modestly significant risk factor for violent behavior. The authors’ analysis showed that 90% of the individuals with a serious mental disorder (i.e., schizophrenia or major affective disorder), excluding substance abuse, were not violent. However, this estimate was based on the subjects’ self-reports. More specifically, because mental disorders are actually rare among the general population, the proportion of violence that may be attributable to mental illness alone was less than 5%. Two questions regarding the ECA data can be posed: 1) Which types of mentally ill people may pose a risk for violence in the community? and 2) What kinds of violence are more likely to occur? Also important are the precipitants of violence and the context in which it occurs. The contaminating factor is the possibility that substance abuse alone could be the variable that increases a person’s relative risk for violent behavior. A high rate of harmful behavior within a community might then come from a “loading” of people who have one of the serious mental illnesses but who primarily act only when under the influence of drugs or alcohol. In Swanson and colleagues’ study, nearly a third of those with schizophrenia or a schizophreniform disorder met criteria for drug abuse or dependence.


103

ASSESSING HOMICIDAL BEHAVIOR IN SCHIZOPHRENIC INDIVIDUALS Before the specifics of homicide and schizophrenia are reviewed, one caveat must be noted. Questions about violence and schizophrenia often elicit extreme reactions from those who believe that such efforts unnecessarily stigmatize a certain group of mentally ill individuals, such as viewing schizophrenia as a neurological disease might do. However, only with further research can the validity of any connection between schizophrenia and violent behavior be confirmed or disconfirmed. It is hoped that accumulated evidence will serve to improve treatment and preventive measures. A person’s dynamics and psychopathology, predisposing factors of a biological and social nature, and current environmental variables all contribute to some final common pathway for a homicide to occur. Too often the complexity of these multifactorial elements is ignored.

Specificity of Studies on Schizophrenic Individuals One important question is whether anything specific can be said about homicidal violence when only the diagnosis of schizophrenia is considered. Studies give estimates that range from 8% to 45% of individuals with a schizophrenia diagnosis showing tendencies toward violence. Various possibilities can account for such a wide variance in these studies. One possibility is that the diagnostic criteria for schizophrenia have changed—despite the use of checklists such as those in the current American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders—and clinicians may differ even on symptoms for checklist inclusion. The shifting of schizoaffective disorder out of the schizophrenia grouping to “psychotic disorders not elsewhere classified” in DSM-III-R27 and then back into “schizophrenia and other psychotic disorders” in DSM-IV 28 and DSM-IV-TR29 may have contributed to some variance in definitions used in the studies. In addition, other types of schizophrenia, such as the disorganized type and especially catatonic variations, may have violent outbursts associated with them that are different from those seen in paranoid schizophrenia. Yet the former varieties of schizophrenia are not investigated for extreme violence nearly to the degree that the paranoid varieties are. However, such people appear fairly routinely in emergency departments and are held in jails. There is also the diagnostic problem of differentiating a diagnosis of delusional disorder. When loose associations are present, the differentiation may not be that complicated, but the individual may not be presenting with that symptom at the time of an evaluation.

104


Whether an act of homicidal violence can be specifically connected to the signs and symptoms of schizophrenia remains a plaguing question. In a Swedish study of violent acts committed by 74 schizophrenic patients, the investigators found that only 36.5% of those acts could actually be related to an active hallucinatory or delusional state.30 One further question that then arises is whether the impaired interpersonal relations of schizophrenic individuals are what primarily influence the disposition toward a violent act. The low-grade interpretive distortions that impinge on schizophrenic individuals’ relationships with others could color events leading to a homicide. Within this framework, the active symptoms and signs may not then always be the sine qua non leading to a homicide. This consideration makes interpretations of the hows and whys of a homicide committed by a schizophrenic person perhaps more complicated to make, but at least potentially closer to the way such events actually unfolded. Opinions vary widely as to what proportion of those in a criminal justice population have “pure” schizophrenia that is not mixed in with comorbid diagnoses. It is prudent to keep in mind that schizophrenic individuals, like those with other major mental disorders, have an increased risk for substance use and abuse.31 Schizophrenic individuals also have an increased risk for personality disorders mixed with substance abuse problems.32,33 As noted earlier, when a schizophrenic person commits a homicide, the question always arises whether the act is attributable primarily to schizophrenia, to the person’s drug or alcohol disorder, to a personality disorder, or to a complex amalgam of all these problems. Alternatively, none of these factors may have been the primary factor leading to the homicide, but instead factors relating to some other basis for the person’s behavior may have been operational. Just because these diagnostic problems exist in a person does not mean that they always explain why a killing has occurred. It is intriguing to note that in studies from Sweden and Denmark to Canada, about half of the homicide offenders with diagnoses of major mental disorders did not have any additional substance abuse diagnoses.34–36 However, these countries have low rates of homicide compared with the United States, as well as perhaps lower rates of substance abuse, so the findings may not be similar in this country. Yet another approach has been to differentiate a high-violence group from a low-violence group based on neurological findings, such as impairments in coordination, tandem walking, hopping, stereognosis, and graphesthesia.37 Just as with other approaches, specific neurological variables need to be assessed in efforts to determine whether people with schizophrenia are more subject to violence than those in other


105

groups. The presence of certain organic findings, if confirmed in schizophrenic individuals, would then raise new questions as to whether these constitute the predisposing diathesis toward homicide. Another subtle difference that can create discrepancies in study results is which subjects have been studied, and when. For example, one study began with a group of individuals who had committed homicide but had already been designated as mentally ill with accompanying diagnoses. In this group, 57% of the murderers were given a diagnosis of schizophrenia.38 Using such a skewed sample gives a quite different result than if a general population of nonmentally ill murderers were used. Other investigators have focused on specific kinds of murder, such as sexual or serial murders. Even apart from basic methodological principles of adequate sample size and the means of investigation, the incidence of schizophrenia may also vary as a correlate of what type of homicide is being studied. In a Canadian study of parricides and attempted parricides, it was noted that parricides represent 3.7% of all homicides in Canada but that parricides are 20%–30% of the homicides committed by psychotic subjects, with more than half having the diagnosis of paranoid schizophrenia (56%) and another 13% having schizoaffective disorder.39 The role of stalking behavior in those with an unfolding thought disorder also needs to be considered. Many of the studies have focused on such behavior when directed at conspicuous targets, such as political figures or movie or sports celebrities. However, stalking with a homicide potential occurs in the more mundane areas of life as well.40 Some stalkers are schizophrenic, although certainly not all stalkers are. It seems surprising that the stalkers who approach their targets do not seem to be the ones who commit actual acts of violence.41 However, psychiatrists are just starting to acquire more knowledge about this subgroup. The specific question of whether those with a diagnosis of schizophrenia have a higher incidence of homicidal behavior remains extremely difficult to answer. As noted, many of the studies have been done outside the United States, and they have usually addressed the outcome of violence rather than homicide per se. Many studies of violence use different definitions for this behavior, which leads to conclusions that are not comparable—and often contrary, as the following examples show: Baxter42 found that those diagnosed with schizophrenia had higher rates of violence than a comparison group of those with other Axis I diagnoses. Another study found females with schizophrenia had higher rates of violence compared with women with other diagnoses, but the finding did not hold for males with a diagnosis of

106


schizophrenia.43 Yet another group found people with schizophrenia had a lower rate of violence compared with those with a different Axis I diagnosis.44 A Swedish study came out with findings of no significant difference in the incidence of violence among persons with Axis I major mental disorders.45 Other studies have found most homicides connected to Axis II diagnoses rather than Axis I.46,47 A caveat is that most studies focusing on some type of mental disorder diagnosis may miss that the incidence of violence in such groups may be higher than in a random sample taken from the population. The MacArthur Study of Mental Disorder and Violence48 defined violence as when a patient placed another person in danger or decided to take some physical action against another person. One of the findings was that the most frequent primary research diagnosis was depression (41.9%), followed by substance abuse/dependence (21.8%), schizophrenia (17.0%), and bipolar disorder (14.1%). The prevalence of violence was divided into three broad groups: major mental disorder, no substance abuse; major mental disorder and co-occurring substance abuse/ dependence; and other mental disorder and substance abuse. The last group was composed of those with a personality or adjustment disorder or cases of “suicidality” and a co-occurring substance abuse/ dependence disorder. The findings after 1 year on the occurrence of violence were 17.9%, 31.1%, and 43.0%, respectively. In terms of specific diagnoses, the 1-year prevalence of violence was 14.9% for patients with schizophrenia, 28.5% for patients with depression, and 22.0% for patients with bipolar disorder. If there was a co-occurring diagnosis of alcohol or drug abuse/dependence with one of the mental disorders discussed earlier, the comparative violence went up from 10.0% to 22.5% in 20 weeks when compared with a group without the added feature of alcohol or drug problems, and within 1 year the difference was 18.1% compared with 31.3%. The presence of a co-occurring diagnosis of substance abuse or dependence was then a key factor in the occurrence of violence. A second finding was that violence was more prevalent with a diagnosis of depression than with a diagnosis of schizophrenia. A warning from such research is that patients cannot be viewed as a homogeneous group when conducting research on violence. Important differences may be concealed and the factor of personality disorders not given adequate consideration. Since 1980, when borderline personality disorder was recognized in DSM-III,49 the potential for violence in those who carry this diagnosis with micropsychotic episodes has always been a possibility. When the factor of substance dependence is added, the potential for violence is aggravated.


107

Delusions, Hallucinations, and Violence Delusions—false beliefs firmly maintained despite negative evidence of social reality—are a challenging problem for assessments of homicidal levels of violence. This difficulty exists not because experienced clinicians have not had cases in which they could demonstrate the connection of such beliefs to violence. Rather, the presence of delusions does not guarantee an outcome of violence—or even that the individual will act on that belief, even though delusional beliefs are a hallmark of a psychotic condition. Thus, delusions are a disorder of thinking that interferes with reality testing through misinterpretations: a part of the person’s reality cannot be validated and thus shared with others. The wide variety of types of delusions is not pursued here, and the focus is rather on illustrating how the presence of delusional thinking is one category of psychosis that may contribute to violence, albeit in a statistical sense; however, only a minority of individuals with delusions commit violence. An intriguing question is why some delusions lead to acts of violence. Persecutory delusions seem more likely to be acted on than other types of delusions.50 Individuals with delusions who have exhibited violence confirm that at least one violent incident was connected to a delusional belief.51 In the gray area of delusion are individuals who have a suspicious nature that leads them to misinterpret the actions of others and who may think, for example, that others have a hostile intent toward them. Clinicians may interpret this idea as something short of a delusional belief, although these individuals’ misinterpretations may also lead to violence.52 An original and creative hypothesis was put forth by Link and colleagues to try to answer the question of why some delusional beliefs produce violent behavior.53,54 The idea was referred to as a threat/control override, which hypothesized a set of beliefs that others wished to harm the person and gain control of his or her mind. These persons then believed that their minds were dominated by forces not under their control, with the experience being that of fear combined with passivity. Given such an experiential state, these persons then became more likely to take some action to escape the feeling of losing control. However, the MacArthur Study of Mental Disorder and Violence was not able to confirm this hypothesis.55 A general group with delusions was followed up at 20 weeks and at 1 year and did not show a significant relationship with violence; group members even showed a weak negative relationship with violence at the end of the year—that is, they were less likely to be violent after discharge. This finding raises a question of whether the treatment received could have ameliorated the violence potential.

108


Interestingly, studies when hallucinations are present also fail to confirm an increased likelihood of violence. An initial problem is that hallucinations, in contrast to delusions, are not nearly as pathognomonic of a psychotic state. Situations of increased stress and various personality disorders can also elicit hallucinations. A similar problem arises in terms of assessing homicidal risk in the presence of hallucinations. Even when command hallucinations of the type “Go kill” were present, a positive relationship was lacking between the hallucinated commands and violence.55 The MacArthur study also did not find that hallucinations in general, or command hallucinations specifically, were significantly associated with violence in the first 20 weeks or over 1 year’s time. However, if voices commanded these individuals to commit acts of violence against people, those individuals were more likely to be violent at both time intervals.

Case Example Clinical material is always valuable to illustrate how difficult it may be to assess violence in a psychotic delusional person. The historic case of Ernst Wagner is a classic from the realm of paranoia.56 Wagner harbored delusional ideas for 12 years without anyone knowing about his violent preoccupations, which eventually erupted in a mass homicide in 1913— first of his wife and four children and then of several others in a town where he had lived. During the 4 years preceding the homicides, Wagner kept a diary amounting to three volumes of his thinking. The diaries reveal that 4 years earlier Wagner had worked out a plan for murder and arson with precise detail. The diary set down his innermost thoughts, which allow us to see the unfolding of his mental illness, in contrast to so many cases where the person, after a homicide, is unable or unwilling to provide information, even if allowed to do so by an attorney. In many cases, the person lacks insight into the possibility that he or she may be ill. Wagner was an esteemed schoolteacher in Germany, and 3 years before the homicides he was recommended for promotion. The evening before the homicides of his wife and children, he had engaged in polite conversation with a teacher’s wife and daughter. (This behavior often characterizes delusional homicide cases and later is used in court, where lay witnesses are introduced to testify how normal the person appeared shortly before the homicides, to try to illustrate the lack of insanity.) The homicides occurred on September 2–4, 1913, when Wagner was 39 years old. He first killed his wife and then their four children by cutting their carotid arteries while they slept. He then left his hometown and traveled by train and bicycle to a village where he started four fires during the night. When people ran out of their houses, he shot at the male inhabitants, killing 8 people and wounding 12 others. This plan had been “hammered” into his brain for several years as revealed in the diary. The day before, he had sent letters to relatives and friends, and a


109

postcard to his landlady asking forgiveness for the unpleasantness that his deeds would cause in her house. The materials show he intended to suicide after the acts. After the homicides, there was endless conjecture in the press. The acts seemed incomprehensible. Everyone remembered Wagner as a teacher and respected citizen. As a teacher, he had been noted for his grand manner and stilted way of speaking and for a “fanatical love of truth.” Wagner admitted what he had done, and the only regret he expressed was that he had been prevented from completing his whole plan. His last letters hinted at indecent conduct. At age 18, Wagner began to masturbate and experienced anxiety about this, despite reassurances from his physician. He became certain that others knew about his secret sins by his appearance. At age 27, on his way home after drinking, a key event occurred. The specifics are not known, but the diary refers to “acts of indecency with animals.” No one knew anything of these acts until they were mentioned in the letters sent before the homicides. In his private thoughts, he had great shame and a feeling that he had sinned against all humankind. He thought that people knew of his disgraceful behavior and were making remarks about him as well as jeering and mocking him. At examination time, he began to carry a loaded revolver in case the police came for him, so he could shoot himself. At his wedding, he had carried two revolvers. Lacking the courage to suicide increased his self-loathing. Diary entries for 1909 reveal the exact plan that he would enact in 1913. Over the next 4 years, he shrank from killing his family and expressed his rage in writing poetry. Feeling tormented, he obtained a school transfer but found no personal relief from his thoughts that bestiality was worse than murder. He continued to have ideas of reference that he was being mocked, and in 1908 the idea of suicide changed to a plan for revenge. All this time, he continued to function as a teacher and was seen as an intelligent person with accurate judgment. In his plan, his hatred would be enacted against the village where he believed he was mocked, and the idea was that he would burn the houses and raze the village to the ground. As he put it, “One hates the places where one has gone astray, the stone on which one’s foot has stumbled.” A question could be raised as to how his delusional beliefs, about being tormented and persecuted by townspeople due to his acts from 12 years earlier, were related to killing his wife and children. Some would argue that killing his family was not directly related to the delusions. The broader position is that killing his family could not be separated from the years of feeling persecuted for his bestial acts and feeling that his family would have been left defenseless and persecuted in a world full of his evil enemies. As a postscript, no public trial was held, but on the basis of two independent psychiatric reports, Wagner spent the remainder of his life in an asylum, asking to be executed. He died in 1938 of pulmonary tuberculosis. Over the intervening 25 years, his delusional beliefs varied in in-

110


tensity but persisted with ideas of reference and persecution, with a deep pessimism and contempt for others remaining with him to the end. He had tried unsuccessfully to have his case reopened legally so he could prove he was mentally healthy and then be sentenced to death.

State of Mind When Committing Crime One controversial issue has been whether violence occurs more frequently in those with a schizophrenic illness during psychotic episodes or when they are free of psychotic states. Differences in schizophrenic individuals’ potential for dangerousness among studies may well reflect the variable of when those individuals are seen. If the individuals are evaluated at an early period when acutely disturbed, the conclusions reached in the assessment may be quite different than if the subjects were studied after a period of treatment and hospitalization. The typical example is an acutely paranoid schizophrenic individual who responds quite readily to neuroleptic medication. If the same person is seen only a few weeks later, a very different clinical picture may emerge with respect to an assessment of homicidal dangerousness. The point cannot be emphasized strongly enough that transposing the results of studies done in inpatient settings to the schizophrenic population in communities is highly unreliable. A classic study was done in 1967 by two famous criminologists, Wolfgang and Ferracuti.57 In their analysis of the literature dealing with psychiatric aspects of homicide, they did not find that any specific diagnosis was connected to perpetration of a homicide. They then concluded that few murderers were psychotic. Yet based on our current knowledge, questions about these conclusions can be raised. How valid were the diagnostic studies carried out on groups of murderers in the criminal justice system in the 1960s? Current assessment techniques have more specificity in detecting disorders in such individuals. Diagnosis has achieved greater levels of sensitivity and specificity even on a purely descriptive level. Unfortunately, few past or present studies devote the time and resources to reconstruct what the mental state of individuals was at the time of committing the homicides, unless a mental illness defense has been contemplated legally and an expert team of lawyers and psychiatrists has been able to study the case intensively. There is no standard methodology for the retrospective determination of an individual’s mental state. It has been noted that research has attempted to address the reliability and validity of criminal responsibility evaluations, but the comparability and generalizability are limited. Reliability evaluations of mental states at the time of an offense are rare,


111

and good validity studies are practically nonexistent.58 Yet the mental state at the time of a homicide is the crucial time period legally. To get away from juries simply trying to decide based on how believable an expert may seem, courts have moved in the direction of creating legal standards based on the reasonableness of methodology employed and the reliability of methods used.59 One approach to assessing the influence of psychotic symptoms on criminal behavior was that used by Rogers and colleagues,60 who differentiated “high-visibility” patients from “low-visibility” patients; the distinction was based on the high-visibility patients’ displaying behavior such as verbal threats, whereas the low-visibility patients did not. The goal was to correlate different psychopathological characteristics in each type of patient. The high-visibility group scored higher on the hostilitysuspiciousness factor of the Brief Psychiatric Rating Scale, whereas the low-visibility group scored higher on the withdrawal-retardation factor of that scale. Certain symptoms suggested a higher potential for serious violence, such as the presence of delusions, especially well-planned paranoid ones. Violence occurring in a disorganized psychotic state, such as in undifferentiated schizophrenia, was thought to pose less of a threat from its being poorly focused, although its unpredictability could have ramifications. The authors also found that symptom pictures involving command hallucinations as well as more exotic syndromes such as erotomania or Capgras’ syndrome could increase an individual’s tendency toward committing homicide. The acutely psychotic individual with thought fragmentation, hallucinations, and misinterpreting would be seen as posing a higher danger. When investigations are extended beyond the descriptive level to address the subtleties of whether an altered mental state was present at the time of the killing, questions arise about the degree of regression or disorganization that was present, from which one could infer a psychotic state. These are different types of questions using different criteria to address inferential subtleties about a person’s mental state and cognitive processes. Few studies, beyond a few intensive legal cases, have carried out the intensive assessments necessary. If such intensive evaluations were done, it would be necessary to distinguish whether differences occurred among the various types of homicides. Distinctions would need to be made among different types of killers, such as sexual killers and serial killers, and among different psychiatric diagnoses. Regarding schizophrenic individuals, distinctions would need to be made among those who exhibited episodic dyscontrol, killed sadistically, dismembered their victims, or murdered children; children who murdered their parents; persons who killed their spouses; and the im-

112


pulsive type of killers once referred to as “sudden murderers.”61 For depressive conditions, the presence of psychotic features with delusions is seen as relatively frequent in the general population, affecting 4 in 1,000 individuals.62 This subgroup of depressed individuals would also raise the potential for homicidal behavior. Similarly, studies show a broad pattern of cognitive impairments in bipolar patients that increases the possibility of their acting on these deficits, with an attendant increased homicidal risk.63 Thus, many of the types of murderers discussed in this book would require psychiatric investigation, in which the possibility that their mental state was changed at the time of the killings would have to be entertained. Their system of controls could be seen as faltering at the time because of a psychosis, some variant of schizophrenia, or a schizophreniform state. Although the incidence of schizophrenia appears similar in different countries and cross-culturally, the incidence of violence actually varies for individuals given a diagnosis of schizophrenia. The type of cross-cultural situation in itself merits a more extended inquiry. Within the designated group of mentally ill murderers, the most frequent single diagnosis given is usually paranoid schizophrenia. Yet even with diverse populations, that diagnosis is given in only about 25% of cases. Only in one study did the diagnosis reach 40%.64

Legal Issues A diagnosis of some type of psychosis in someone who has committed a homicide often raises major legal issues involving an insanity defense. Initial issues of a person’s competency to participate in a trial may surface; such a ruling would negate the possibility, at least for a time, of an individual standing trial. These initial issues would include clearly establishing the proper diagnosis of schizophrenia, because an accurate diagnosis is necessary for any legal or clinical planning. To make the diagnosis, several signs and symptoms and diagnostic tests can be used to confirm or rule out such conditions (none of which are addressed here). In essence, clinical material from the patient is used, as well as whatever extraneous sources are available about that person’s behavior and how the homicide took place. These sources may include information from relatives (e.g., spouse, parent) or friends as well as medical records from treating physicians or past hospitalizations—psychiatric and other—that would help establish the diagnosis and behavioral patterns. Unless such information is incorporated into an evaluation, later research relying on the listed diagnosis is questionable.


113

A contemporary examination of the individual, focusing on his or her current level of functioning, is also called for. This examination involves general impressions of how the person conducts himself or herself, the pattern and content of his or her speech, the types of thought processes and content that have preoccupied him or her, the kinds of moods and affect that have predominated, and his or her style of relating to others and to himself or herself. In addition, a traditional sensorium examination may be useful in some cases to screen out factors such as cognitive deficits in orientation to time, place, and person; to determine the ability to abstract and assess cognitive functioning; to assess the capacity for remote and recent memory; and to evaluate the usual material incorporated in mental status examinations involving the capacity to perform different types of calculations, interpret proverbs, and so on. Psychological testing for personality functioning and an intellectual assessment are routine. As a general rule, it is wise to obtain a general physical and possibly a neurological examination if no recent one has been performed. All of these evaluations are used to determine if the primary signs and symptoms of some type of psychotic disorder are present and how they might be related to a homicide. The following case illustrates the unfolding of homicidal violence in a schizophrenic individual.

Case Example A 40-year-old single male was charged with the murder of his 70-yearold mother. An interesting phenomenon was that despite an IQ of 138 based on the Wechsler Adult Intelligence Scale—Revised, the man had worked most of his life at a newsstand in the central part of a major metropolitan area. Over the years he had become known to the many people who bought their magazines or newspapers from him. No one suspected his superior intelligence, which was undoubtedly far above that of a great majority of his newsstand patrons, because he always dressed shabbily, was usually friendly in a quiet and unassuming way, and kept to himself. Although he had not graduated from high school, he frequently spent his spare time assembling and disassembling complex stereo and computer equipment in addition to reading heavily intellectual books. He had always lived with both parents until 10 years before the homicide, when his father died; after that he continued to reside with his mother. In the course of his interview, what appeared most striking was his expression of his suffering and anguish resulting from what he felt he had been exposed to over the preceding several years. It had begun when he felt himself becoming fragmented because of experiencing extraneous intrusions into him of some type of electrical current or electrical rays. These not only influenced him but were experienced as

114


“invading and assaulting” him. He did not understand the source of these intrusions until shortly before he assaulted and killed his mother, when he first began to associate the events as somehow connected with her. Although he had slowly begun to drink more alcohol at the times when his auditory hallucinations increased in intensity, he was not otherwise a drinker. He referred to the electrical influences operating on him as “ghosts” or “spirits” that interfered with his life, and the drinking provided some relief from them. A crucial transition occurred when he tried to synthesize what was responsible for the events that were happening to him. In contrast to the success he had in unraveling stereos and computers, his efforts failed to unravel the sources of electrical intrusions on himself. The idea emerged that something like ghosts were responsible for putting the strange thoughts in his head or for making him hear the voices directed to him that caused him to become very agitated and hurried in his behavior. A startling change occurred when he began to experience voices coming out of radios or television sets, even when they were turned off. He would then disassemble the sets, but voices kept coming from the parts. His first thought was that some religious group might be behind the persecution, tormenting him because of his disavowal of religion and for leaving his church years ago. He next began to feel vulnerable to public exposure at the newsstand, because he was so visible there to whatever or whoever was persecuting him. He began to feel uneasy and became extravigilant. At times he would simply leave the newsstand and walk around the block, particularly when the voices became more pressing. At those times, he experienced electrical currents coming from street lampposts when he walked by them, and he would hurry back to his newsstand. Riding on buses began to bother him because he heard voices that he connected with the electrical systems used on the bus. As a result, he began to walk rather than use public transportation. When voices shifted to threats against his life, he became acutely agitated. They never said how they were going to kill him, but he became convinced his life was in jeopardy. While at home, he became frightened and began bolting doors and adding extra locks as well as pushing chairs against the doors at night. In some undefined manner, he began to suspect that his mother was the culprit behind these events. Thinking that his mother might be planning to kill him to get his money, he changed ownership of his government bonds to revert them to the government in event of his death. However, that gave him no relief. A new symptom became obvious when he looked at immobile objects, such as a fist or sharp object (e.g., pen, razor). He would feel sharp pain when he experienced the fist as hitting him or the razor as cutting him. On one occasion, he looked at a nail file and felt it was penetrating his eyes to the point of blinding him. He became unable to distinguish whether events that were reported in the news had special significance for him, such as events involving celebrities or a criminal event such as a local robbery. Personalistic use of phrases began to occur in his auditory hallucinations. A voice would say, “We’re going to dehand you,”


115

which meant to him that his hand was going to be removed. Rhyming and punning of words occurred, such as hearing the word “devacilate,” which he interpreted as referring to a procedure to remove his veins, or “demasculate,” which he interpreted as a procedure to remove his organs and nerves. The culmination of his delusions came on a night when he became fully convinced that if his mother was not one of the main perpetrators of the misery being inflicted on him, she was at least aware of what was going on and was doing nothing to help him. He concluded that his mother was allowing whatever was in store for him to happen, whether it be pain, suffering, punishment, even perhaps death. He had one vivid memory of standing in a doorway between the living and dining rooms, seeing a pinkish white spot on his mother’s cheek below her left eye that looked like a “patch of flesh” in her left eye socket, and figuring that her eyeball had come out. “I seemed confused. I didn’t know if she was dead or alive, and I thought, ‘Oh! Oh! We had an argument and I must have hit her.’ ” At the same time, he was frightened of angering his mother, because she might ask him to move out of the house. “I was afraid that if I had to leave home, how could I plug up all the electrical outlets or shut off the electricity in other places like I had done at home, because they wouldn’t allow me to do that.” Although he could not remember more details of the events of that evening, the police reports and the autopsy on the mother indicated that her head had been beaten so severely by the leg of a chair that had been found near her body that no visible sign of a head being attached to her body remained. He recalled that once during the night, he had come down from his bedroom and noted, “My mother’s head looked like it was dissolved from acid and not pleasant. The bone and everything else looked like it had been dissolved—as though acid had been spilled on it. I knew then that she was dead.” At that point, he called the police.

In many ways, this fascinating and ghoulish case emphasizes the importance of clearly establishing a diagnosis and the validity of that diagnosis. Clinical issues must then be considered to inform legal decisions about whether an individual is able to stand trial and, if so, whether he or she is mentally ill to the degree of meeting the prerequisite insanity standards in that jurisdiction. In this case example, the defendant did not appear to know that killing his mother was wrong, given the delusional context in which the killing occurred. Once he became convinced that his mother was, in some manner, involved in the life experiences he was having, her life was in serious danger. Unless there had been significant therapeutic intervention, or it was found that these psychotic episodes would come and go, the possibility of a gradually unfolding and expanding psychotic process was present. The defendant’s belief that he was involved with the victim in some type of persecutory system put him in the role of logically believing

116


he needed to protect himself. The ideation that the electrical forces from outlets and equipment, as well as from outer space, were causing him difficulty; the form of his physical symptoms; and his belief that he was being attacked at first centered around his home, then expanded to his job site, and finally led him to conclude that his mother was the persecutor. All these thoughts coalesced on the night of the killing. He began to fear that his mother would throw him out of the house and he would not be able to maintain his elaborate defenses to hold the electrical systems in abeyance. It is not unusual with such psychotic episodes that the specific details of a significant act of violence are not remembered. Instead of this not remembering being an attempt on the part of the perpetrator to dissimulate, it is a type of posthomicidal reparation that occurs subsequent to a psychotic fragmentation in which a homicide has occurred. This type of case may produce a multitude of legal issues involving a defendant’s competency before coming to trial or the state of his or her mind at the time of the killing. Although the case example does not present unusual clinical difficulty in diagnosis, it is a good one to illustrate related problems. Many encountering the case might expect that it would easily fit into a legal niche of insanity, as is often the case. However, it need not be so, because many variables—legal and political—impinge on insanity dispositions. One argument that might be raised within a traditional framework of legal tests is whether, because of his mental illness operating at the time of the killing, this defendant knew the nature of his violent act. A relevant question is whether he knew he was factually destroying his mother or whether he perceived her at the time as some robotized object or a computerized source of audio messages directed at him. However, if he knew his act was wrong, did he believe it was a necessary act to eliminate the source of his persecutions? From a strict prosecutorial position, many of these arguments might be objectionable because even though the person was psychotic and delusional, the psychotic framework did not necessarily mean the person was under an immediate fear for his life because of his delusions and hallucinations, despite his personal anguish. As a consequence, the prosecutor’s position would be that the extreme solution this defendant took was not proportional to the threat facing him, even accepting his delusional belief. In other words, his being convinced that he was being tormented within a delusional system might only justify an equivalent response in return, which would be something short of a killing. These types of cases could lead to an elaborate discussion of the insanity defense and its vicissitudes, but the point here is simply to illustrate how legal uncertainties can arise even when a schizophrenic person kills someone in the midst of a psychotic episode.


117

REFERENCES 1. Pescosolido BA, Monahan J, Link BG, et al: The public’s view of the competence, dangerousness, and need for legal coercion of persons with mental health problems. Am J Public Health 89:1339–1345, 1999 2. Group for the Advancement of Psychiatry: Psychiatry and Sex Psychopath Legislation: The 30s to the 80s. New York, Group for the Advancement of Psychiatry, 1977 3. Lamb HR: The Homeless Mentally Ill. Washington, DC, American Psychiatric Press, 1984 4. Isaac RJ, Armat VC: Madness in the Streets. New York, Free Press, 1980 5. Felthous AR: Duty to warn or protect: current status for psychiatrists. Psychiatr Ann 21:591–597, 1991 6. Tarasoff v Regents of the University of California, 17 Cal 3d 425, 551 P2d 334, 131 Cal Rptr 14 (1976) 7. Stone A: Law, Psychiatry and Morality. Washington, DC, American Psychiatric Press, 1984 8. Livermore JM, Malmquist CP, Meehl PE: On the justifications of civil commitment. University of Pennsylvania Law Review 117:75–96, 1968 9. Monahan J: The prediction of violent behavior: toward a second generation of theory and policy. Am J Psychiatry 141:10–15, 1984 10. Dawes RM, Faust D, Meehl PE: Clinical versus actuarial judgment. Science 243:1668–1675, 1989 11. Lowenstein M, Binder RL, McNeil DE: The relationship between admission symptoms and hospital assaults. Hosp Community Psychiatry 41:311–313, 1990 12. Beck JC, White KA, Gage B: Emergency psychiatric assessments of violence. Am J Psychiatry 148:1562–1565, 1991 13. Stark R, McEvoy J III: Middle class violence, in Violence in the Family. Edited by Steinmetz SK, Strauss MA. New York, Harper & Row, 1974 14. Barnett O, Miller-Perrin CL, Perrin RD: Family Violence Across the Lifespan, 2nd Edition. Thousand Oaks, CA, Sage, 2005 15. Ashley MC: Outcome of 1000 cases paroled from the Middletown State Homeopathic Hospital. State Hospital Quarterly 8:64–70, 1922 16. Cohen LH, Freeman H: How dangerous to the community are state hospital patients? Connecticut State Medical Journal 9:697–699, 1945 17. Hastings DW: Follow-up results in psychiatric illness. Am J Psychiatry 144:1057–1066, 1958 18. Gulevich GD, Bourne PG: Mental illness and violence, in Violence and the Struggle for Existence. Edited by Daniels DN, Gilula MF, Ochberg FM. Boston, MA, Little, Brown, 1970, pp 309–326 19. Rappeport JR, Lassen G: Dangerousness—arrest rate comparisons of discharged patients and the general population. Am J Psychiatry 121:776–783, 1965

118


20. Rappeport JR, Lassen G: The dangerousness of female patients: a comparison of the arrest rate of discharged psychiatric patients and the general population. Am J Psychiatry 123:413–419, 1966 21. Giovannoni JM, Gurel L: Socially disruptive behavior of ex-mental patients. Arch Gen Psychiatry 17:146–153, 1967 22. Sosowsky L: Crime and violence among mental patients reconsidered in view of the new legal relationship between the state and the mentally ill. Am J Psychiatry 135:33–42, 1978 23. Guze S: Criminality and Psychiatric Disorders. New York, Oxford University Press, 1976 24. Cloninger CR, Guze SB: Psychiatric disorders and criminal recidivism. Arch Gen Psychiatry 29:266–269, 1973 25. Zitrin A, Hardesty AS, Burdock ET, et al: Crime and violence among mental patients. Am J Psychiatry 130:144–146, 1976 26. Swanson JW, Holzer CE III, Ganju VK, et al: Violence and psychiatric disorder in the community: evidence from the Epidemiologic Catchment Area surveys. Hosp Community Psychiatry 41:761–770, 1990 27. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987 28. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 29. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 30. Virkkunen M: Observations of violence in schizophrenia. Acta Psychiatr Scand 50:145–151, 1974 31. Mueser KT, Yarnold PR, Levinson DF, et al: Prevalence of substance abuse in schizophrenia: demographic and clinical correlates. Schizophr Bull 16:31–56, 1990 32. Goodwin FK, Jamison KR: Manic Depressive Illness. New York, Oxford University Press, 1990 33. Rice ME, Harris GT: Psychopathy, schizophrenia, alcohol abuse, and violent recidivism. Int J Law Psychiatry 18:332–342, 1995 34. Lindquist P: Criminal homicide in northern Sweden. Int J Law Psychiatry 8:19–37, 1989 35. Gottlieb P, Gabrielsen G, Kramp P: Psychotic homicides in Copenhagen from 1959 to 1983. Acta Psychiatr Scand 76:285–292, 1987 36. Beaudoin MN, Hodgins S: Homicide, schizophrenia and substance abuse or dependency. Can J Psychiatry 38:541–546, 1993 37. Krakowski MI, Convit A, Jaeger J, et al: Neurological impairments in violent schizophrenic inpatients. Am J Psychiatry 146:849–853, 1989 38. McKnight CK, Mohr JW, Quinsey RE, et al: Mental illness and homicide. Can Psychiatr Assoc J 11:91–98, 1966


119

39. Marleau JD, Millaud F, Auclair N: A comparison of parricide and attempted parricide: a study of 39 psychotic adults. Int J Law Psychiatry 26:269–279, 2003 40. Mullen PE, Pathe M, Purcell R: Stalkers and Their Victims. Cambridge, UK, Cambridge University Press, 2000 41. Deetz PE, Matthews DP, Von Duyne C, et al: Threatening and otherwise inappropriate letters to Hollywood celebrities. J Forensic Sci 36:185–209, 1991 42. Baxter R: Violence in schizophrenia and the syndrome of disorganization. Crim Behav Ment Health 7:131–139, 1997 43. Wessely S: The epidemiology of crime, violence and schizophrenia. Br J Psychiatry 170:8–11, 1997 44. Gardner W, Lidz C, Mulvey EP, et al: Clinical versus actuarial predictions of violence of patients with mental illnesses. J Consult Clin Psychol 64:602– 609, 1996 45. Belfrage H: A ten-year follow-up of criminality in Stockholm mental patients. Br J Criminol 38:145–155, 1998 46. Tardiff K, Marzuk PM, Leon AC, et al: A prospective study of violence by psychiatric patients after hospital discharge. Psychiatr Serv 48:678–681, 1997 47. Wallace C, Mullen P, Burgess P, et al: Serious criminal offending and mental disorder: case linkage study. Br J Psychiatry 172:477–484, 1998 48. Monahan J, Steadman HJ, Silver E, et al: Rethinking Risk Assessment: The MacArthur Study of Mental Disorder and Violence. New York, Oxford University Press, 2001 49. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC, American Psychiatric Association, 1980 50. Wessely S, Buchanan A, Reed A, et al: Acting on delusions, I: prevalence. Br J Psychiatry 163:69–76, 1993 51. Junginger J, Parks-Levy J, McGuire L: Delusions and symptom-consistent violence. Psychiatr Serv 49:218–220, 1998 52. Arseneault L, Moffitt TE, Caspi A, et al: Mental disorders and violence in a total birth cohort: results from the Dunedin study. Arch Gen Psychiatry 57:979–986, 2000 53. Link B, Stueve A: Psychotic symptoms and the violent/illegal behavior of mental patients compared to community controls, in Violence and Mental Disorder: Developments in Risk Assessment. Edited by Monahan J, Steadman H. Chicago, IL, University of Chicago Press, 1994, pp 137–159 54. Link B, Monahan J, Stueve A, et al: Real in their consequences: a sociological approach to understanding the association between psychotic symptoms and violence. Am Sociol Rev 64:316–332, 1999 55. Rudnick A: Relation between command hallucinations and dangerous behavior. J Am Acad Psychiatry Law 27:253–257, 1999 56. Gaupp R: The scientific significance of the case of Ernst Wagner, in Themes and Variations in European Psychiatry. Edited by Hirsch SR, Shepherd M. Bristol, UK, John Wright and Sons Ltd, 1974, pp 121–133

120


57. Wolfgang ME, Ferracuti F: The Subculture of Violence. New York, Barnes & Noble, 1967 58. Simon RI: Retrospective assessment of mental states in criminal and civil litigation, in Retrospective Assessment of Mental States in Litigation. Edited by Simon RI, Shuman DS. Washington, DC, American Psychiatric Publishing, 2002, pp 1–20 59. Daubert v Merrell Dow Pharmaceuticals, Inc., 509 US 579 (1993) 60. Rogers R, Gillis JR, Turner RE, et al: The clinical presentation of command hallucinations in a forensic population. Am J Psychiatry 147:1034–1037, 1990 61. Blackmun N, Weiss JM, Lambert JW: The sudden murder: three clues to preventive interaction. Arch Gen Psychiatry 8:289–294, 1963 62. Ohayon MM, Schatzberg AF: Prevalence of depressive episodes with psychotic features in the general population. Am J Psychiatry 159:1855–1861, 2002 63. Bearden CE, Hoffman KM, Cannon TD: The neuropsychology and neuroanatomy of bipolar affective disorder: a critical review. Bipolar Disord 3:106–150, 2001 64. Lunde DT: Murder and Madness. Stanford, CA, Stanford Alumni Association, 1975

CHAPTER

4 BORDERLINE PERSONALITY DISORDER AND HOMICIDE The Quest for Vindication VALIDITY OF AND CRITERIA FOR DIAGNOSIS OF BORDERLINE PERSONALITY DISORDER When the diagnosis of borderline personality disorder (BPD) is a possibility in someone who has committed a homicide, an initial and primary issue that arises is the diagnostic validity of the entity. Although the same question arises with many other diagnoses in terms of a differential diagnosis, for multiple reasons the question arises in a more cogent form with BPD. For the purposes of this chapter, I wish to note that this disorder has been subjected to diverse interpretations, from being viewed as a subtle form of schizophrenia to being conceptualized as a neurotic condition in psychoanalytic theory.1 Use of the diagnostic interview for borderline patients 2 and the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria has improved the validity of diagnosing this disorder. However, questions remain as to whether the diagnosis, officially recognized only since 1980 in DSM-III,3 has yet been adequately conceptualized. The problem is attributed to the mixture of fluctuating symptoms and behaviors listed, which comprises both states and traits. It has been noted that because only five of the eight criteria listed in DSM-III-R4 need to be present to make the diagnosis, there are 93 different ways to arrive at the diagnosis.5 Different editions of DSM also changed the criteria. As recently as the late 1970s, it was argued that BPD was an independent diagnostic entity, although there was no preponderance of evidence to weigh either for or against this hypothesis.6 Another set of

121

122


questions has arisen as to whether there is a need for a different formulation of this personality disorder, with suggestions that a revised formulation would see the problem more in terms of levels of a disordered self-organization rather than simply symptoms and behaviors.7 Indeed, Kernberg8 argued for three levels of personality disorganization—neurotic, borderline, and psychotic—that would include several descriptive diagnoses from DSM of personality disorders (paranoid, histrionic, dependent, schizoid, schizotypal, narcissistic, borderline and antisocial). He also incorporated personality disorders not in DSM (sadomasochistic, hypochondriacal, cyclothymic, and hypomanic). In addition, a large overlap exists between BPD and other types of personality disorders, which raises questions about the pervasiveness of comorbidity, especially for diagnoses involving drug and alcohol abuse and the spectrum of impulse disorders. Comorbidity with Axis I affective disorders is often noted.10 In fact, it is rare that an individual with BPD is free from a comorbid Axis I type of diagnosis.11 The most common would be some type of substance abuse or dependence that can be a facilitator of violence, with the person as a user or distributor. Other frequent comorbid conditions include eating disorders, posttraumatic stress disorder, dissociative states, and panic attacks. The pervasiveness of mood disorders always raises questions about the overlap of these disorders. This overlap has produced continuing debate as to whether a group of those with BPD actually reflect a subtype of a mood disorder. Finally, other lines of inquiry suggest the possibility that certain kinds of biological dysfunctions are present more frequently in those with the BPD diagnosis,12 such as shortened rapid eye movement latency and sleep continuity disturbances, abnormal dexamethasone-suppression test results, and abnormal thyrotropin-releasing hormone test results. Whether these changes are due to comorbid depression is the question. It has been noted that agreement in diagnosing BPD using different diagnostic instruments hovers at around 50%.13 Although questions about the validity and reliability of a BPD diagnosis are still unanswered, they are listed here simply to indicate that questions may be raised when this diagnosis is considered for a homicidal person or group in clinical, legal, or research arenas. These questions are also mentioned as words of caution before turning to the forensic implications of a BPD diagnosis and its relationship to homicides. Interspersed are qualities of irritability and affective storms that can presage violence, but assessing the possibility of violence remains elusive. When paranoid thinking is mixed with feelings of rage and anger, the violence potential is magnified.

Borderline Personality Disorder and Homicide

123

ETIOLOGY OF BORDERLINE PERSONALITY DISORDER AND PROPENSITY TOWARD VIOLENCE A key question in creating a theoretical formulation about BPD and violence is to ask what special vulnerability exists in the structure of this type of personality that allows a homicidal act to occur. The lack of specificity about the etiology of borderline personality and its functioning is troubling; however, certain formulations and inferences can be made. Most striking in this regard is the significance of the pervasive and enduring hatred in persons with BPD, a hatred that persists and guides a person who superficially appears to be socialized and integrated much of the time. However, multiple theories exist to explain the etiology of BPD.

Early Theories The nineteenth-century ideas about temperament are forerunners of attempts to understand people with BPD. The first approach to explaining those people who displayed anger mixed with moodiness was that they had a choleric temperament, which was marked by a quality of irritableness. Later theorists distinguished such a group as having an innate amount of aggression. Over the ensuing decades, other theories attempted to relate the behavior of this group of people to schizophrenia by way of schizoid elements. By 1968, Grinker and colleagues14 had delineated four subtypes of BPD individuals, and all four subtypes had the core syndrome of acting out of anger. The angry core was seen as the individuals’ connection to their impulsivity and instability. Liebowitz and Klein15 used the term hysteroid dysphoric to refer to impulsive female patients who were chaotic, moody, unreasonable, tempestuous, and irritable. Extreme forms of behavior in this group were witnessed in violent acts where chaotic impulsivity, wildly oscillating extremes of love and hate, and compulsive promiscuity were played out. Given BPD individuals’ manipulative qualities, a number of variations can occur. In one variation, third parties are lured into carrying out acts of violence at the individual’s promotion. For example, one girl told her boyfriend lurid tales about her father molesting her, stories that were never proven true. The boyfriend was then incited to kill her parents.16 Diverse types of distortion, lying, and manipulation are played out to various degrees in “doing someone in,” up to and including committing a homicide. Sometimes homicides involving such behaviors are behind the duplicity present in business dealings or a white-collar crime that has backfired.

124


Biological Theories As noted earlier, various hypotheses have been suggested to explain the source of hatred in those with BPD. Biological theorists have proposed that some intrinsic neurological deficit of brain function or a genetic predisposition contributes to these individuals’ problems with impulse control. There is a scarcity of twin studies dealing with BPD inheritance. A Norwegian twin study suggested that the effect may be larger than anticipated, with the genetic portion of variance approaching 0.70.17 This finding would be consistent with the literature suggesting that personality traits, similar to BPD, exert a genetic influence. Biological theorizing also looks for something deviant in neuropsychological functioning, perhaps hyperirritability, as a precursor to impulsivity. From there, the generalizations extend to hypotheses such as temporal lobe epileptic variants, with violence as an accompaniment; episodic dyscontrol; panic attacks; dissociative syndromes; premenstrual syndrome with destructive behavioral components; and bipolar II disorder or cyclothymia as possibilities corresponding to different manifestations of the borderline group.

Physiological Theories Physiological vulnerabilities in the individual with BPD point to behavior that is impulsive, self-destructive, and suicidal. Impulsivity is viewed as reflecting anger, fear, frustration, and irritability. The neurochemical correlates involve many of the systems discussed in Chapter 2 (“Biological Factors in Homicide”), in the form of four main neurotransmitter systems: 1) noradrenergic; 2) dopaminergic; 3) serotonergic; and 4) γ-aminobutyric acid (GABA).18 Noradrenergic activity is seen as increasing aggressive behavior. Dopaminergic pathways enhance irritative aggression and slow the rate of habituation of aggressive outbursts. Serotonin is seen as inhibiting aggression through the amygdala, which then decreases dopaminergic activity, or perhaps through projections from the medial raphe to the hippocampal area or prefrontal cortex. GABA is also seen as inhibiting aggressive behavior.19 Impulsivity and deficits in planning ahead as an executive function are seen as reflecting abnormalities in the prefrontal cortex.20 However, these behaviors are also seen in other personality disorders as well, such as antisocial personality disorder. These changes may thus be present in different personality disorders and therefore lack specificity. In patients with BPD, an inverse relationship between cerebrospinal fluid 5-hydroxyindoleacetic acid (5-HIAA) and aggression was found.21 Again, this result seems to parallel the find-


125

ings in many personality disorders. Fenfluramine releases serotonin and blocks its reuptake. This effect is tested by examining the prolactin response to fenfluramine; patients with a BPD diagnosis have a blunted response compared with patients with other personality disorders.22 This finding suggests that serotonin responsivity is decreased in aggressive or suicidal patients. The key in this approach to the patient with BPD is trying to understand what triggers the impulsivity and suicidality so often present and the correlation with homicidal levels of violence. Stone23 used an analogy to a neurophysiological model that has either an exaggerated or prolonged response to explain the impulsivity of persons with BPD. The individual may react to less intense provocative situations than others because he or she has a lowered threshold for provocation, a more rapid triggering of a response, an exaggerated response, or a response of extended duration. If one or more of these possibilities can be confirmed, it would not be unexpected to find such people having heightened tendencies for outbursts of impulsive violence. Extreme degrees of the personality organization keep the person in a state of action readiness, with a brooding need to act that can last for months or longer. Such physiological theorizing postulating an inherent instability in the borderline individual is congruent with clinical observations. The implications of such a physiological theory would be congruent with the inherent instability clinically seen in individuals with BPD. Based on features such as a brooding sensitivity with a readiness to perceive slights, the tinderbox potential of exploding in a burst of rage or taking some calculated act of vengeance seems logical given the preceding thinking. Added to this proposed neurophysiological propensity toward impulsivity, developmental experiences, such as diverse types of trauma or abuse, may then contribute to this tendency or cause the impulsivity themselves if severe enough. The difficulty, of course, is that although this hypothesis of inherent physiological instability seems logical, it has not yet been confirmed. Furthermore, no cues predict when a violent eruption may occur. Indicators are lacking to tell us when love will turn to hate, idealization to devaluation, admiration to contempt, loyalty to a person or group to undermining or deceitful behavior, or devotion to envy or vicious jealousy. Even those in close contact with a borderline individual, and not joint participants in his or her psychopathology, will make false negative predictions because of the rapidity of the extreme shifts and impulsivity of such individuals’ actions. The ideas of impulsive aggressiveness and affective instability may individually and in combination set the stage for homicidal behavior.

126


Childhood Theories A persistent question is whether there are some childhood precursors of the adult who will have a borderline personality. Despite the perspective that character traits are a matter of long-term duration that result in stable traits, few studies examine the thread of children with particular traits who will later emerge with a BPD diagnosis. Instead, the focus has been largely on which children with conduct disorders will develop into adults with antisocial personality disorder. In contrast to conduct disorders, two important hallmarks of adults with BPD—unstable intimate relationships and suicidal behavior—are not as available for study in children. Although BPD is more frequent in women with impulsivity and affective instability in contrast to more overt criminality and substance abuse in men, the disorder could come from a common base— with overt exploitation and aggression simply more common in men and with victimization and aggression turned against the self as exaggerations of characteristics more common in women.24 This idea might also account for the predominance of boys being referred to clinics for more overt problems, whereas girls with the potential for BPD problems behave in a more internalizing manner until adolescence, when they become more impulsive. The girls then have a mixture of internalizing and externalizing symptoms. It is interesting that retrospective studies show similar childhood backgrounds in adults diagnosed with antisocial and borderline personality disorder. They come from dysfunctional families, with parents often having serious psychopathology, and overlapping experiences of abuse and neglect.25 The initial idea was to seek childhood precursors of BPD, but this concept yielded little agreement.26 Later the construct of “borderline pathology of childhood” emerged as a complex and severe behavioral syndrome.27 The clinical picture included externalizing, internalizing, and cognitive symptoms. The children were seen as highly impulsive, suicidally depressed, and having some micropsychotic symptoms. These multisymptom impairments are seen as having long-term sequelae and as making up a high-risk group for serious violence. Thus, long-term follow-up of such a group of children is needed with respect to violent outcomes. One study, which referred to the group as “socalled borderline children,” showed a disparate group of disorders by early adulthood.28 However, children with borderline pathology do not become adults with BPD. It is possible—but many outcomes are possible, such as chronic mood disorders. Hence, these borderline phenomena in childhood are not to be conceptualized as an early version of adult BPD. In my experience, the symptoms seen in these children and adolescents were frequently encountered when a homicide occurred.


127

Another perspective in terms of the vulnerability to impulsive behavior has been through investigations of children of borderline parents. Links 29 reported an increased rate of borderline personality in firstdegree relatives of probands. He emphasized how the parents exerted a high degree of hostile control. Children of borderline mothers, compared with control subjects, had significantly more psychiatric diagnoses, more impulse control problems, and a higher frequency of childhood borderline pathology.30 An empirical study of 776 adolescents with personality disorders found the joint presence of maternal overinvolvement and maternal inconsistency could predict the emergence of BPD.31

Psychoanalytic Theories There is a good deal of overlap among childhood theories, psychoanalytic views, and attachment theory regarding the development and ramifications of borderline personalities. Aspects relevant to the potential for a later violent predisposition are selected here. Among psychoanalytic theories, Kernberg’s32 explanatory models have many implications for determining a person’s proneness to violence. His original model posited that an individual’s aggressive drive, emerging in the developmental period between 8 months and 36 months, posed a threat to libidinal object ties. The result of this struggle was defensive efforts to keep aggressive and libidinal urges separate in images of the self and other objects. Such efforts give rise to a defensive splitting and other primitive defenses wherein positive and negative images of the self and others emerge. For those with excess aggression, or difficulties in dealing with aggression, there is difficulty in keeping positive and negative images together in one person. The result can be an impairment in reality testing and exaggerated ambivalence. Adler33 went beyond postulates of a failed integration of self and object representations attributable to ambivalence in his proposal of a developmental failure in the formation of “soothing-holding introjects” during the separation-individuation phase, which occurs during the second and third years. There is a failure to develop stable object constancy in the course of development. The resultant introjects are what can cause harm. The possibility then exists for some future regressive loss of function to the state of recognition memory and reliance on transitional objects. During such regressions, terrifying states of aloneness can occur, with fears and fantasies activated on the level of threats of annihilation of the self or parent figures. In terms of the cases discussed in this chapter, such a primitive ego state can be hypothesized as the state possibly present during homicides.

128


The idea that a fear of being abandoned is central to borderline development overlaps with attachment theory. Masterson’s34 idea was that children’s striving for autonomy led to conflict with a parent who withdrew emotional support at those times. The outcome was a child, and later an adult, who was subject to depressive states and fearful of abandonment if he or she did not continue in a dependent state. The anger and rage that this dependency could later produce are seen in some cases of borderline homicidal behavior. However, it should be noted that traditional psychoanalytic theorists have seen no need for such proposals, believing that classical theory is sufficient to explain the clinical observations about individuals with BPD.35 The ambivalence of the person with BPD is seen as reflecting the anal stage of development, with splitting explained as a displacement of aggression from a loved, caretaking object to preserve the object from a fantasized destruction. Violence would then be seen as a displacement mechanism gone awry. Pursuing a broader theoretical perspective to the problem of BPD and homicidal violence leads to an examination of the vulnerabilities accompanying a borderline personality organization. One vulnerable area has always been the tendency to a diffusion of identity. Researchers have delineated an identity cluster consisting of identity diffusion, empty feelings, and boredom.36 A personality with chronic interpersonal problems in the areas of intimacy, empathy, and assessment of others’ intentions (as well as the person’s own intentions) places that person in a higher risk category for acting on his or her aggression. It also produces greater difficulties in his or her love life, which is the context for many acts of violence and homicide between intimates. Not being able to assess accurately one’s own intentions becomes a greater problem when there are similar difficulties in assessing others’. The potential is then present for acting on misinterpretations and escalating violent behavior. Some of these disturbances appear as hysterical symptoms.37 On other occasions, there may be symptoms of dissociative disorders.38 Grisly types of murders, such as those involving multiple stab wounds or the burning of a dwelling with the victims inside, may occur in a depersonalized state. Sometimes obsessional defenses break down. A woman in her 30s had been seeing a therapist and receiving medication on the basis of her having an obsessive-compulsive disorder. Many of the background factors related to her borderline condition were bypassed. She had recurring thoughts of doing something harmful to others or herself, such as driving her car into a bridge. Apart from cutting herself at times, she had made no other overt physical acts. However, while watching her father use a sledgehammer to do some building, she obsessed about hit-


129

ting him with it. She walked away from the situation several times without doing anything but ultimately returned and hit him on the head, after which she “came to” and called the police.

Attachment Theory Attachment theory overlaps with childhood and psychoanalytic theories, although it stands independently. The original theory is often attributed to John Bowlby,39 who emphasized the need for affectional and reciprocal bonding as an essential component for human development in infancy and thereafter. A variety of responses viewed either as insecure attachments or attachment disorders has been studied in the child development literature. Some of the responses in infants have overtones for later psychopathology, such as infants who do not attach, have a disordered attachment, or experience a loss. These infants may respond to stressful situations with undirected behavior including freezing, head banging, clapping, and trying to escape unfamiliar situations. These behaviors have been described as indications of disorganization or disorientation.40 Although psychoanalytic theory made contributions to attachment theory, it operated outside a biological framework and held to a view that attachment must occur in the oral stage of psychosexual development.41 The focus was also on individual ego development rather than on the interpersonal relationships seen as important in attachment theory. It would be reductionistic to offer an explanation for the violence potential in borderline personalities by attributing it to nothing but the insecurities of childhood attachments and connected anxieties—not because these insecurities are irrelevant but because again there is a lack of specificity. One liability in those with BPD is the difficulty integrating the memory of an experience with the meaning of it.42 BPD individuals may deny memories and then idealize or devalue early relationships. Perhaps the lack of open discussions of emotions in the developmental background of a BPD individual or, more damaging, the prohibition or punishment of emotional expressions does the damage.43 The child is then prone to misidentify verbal communications and emotional cues in others, which contributes to dissociation and impairs integration of experiences. Preoccupations contribute to confusion and anger in interpersonal relationships. The not infrequent history of past maltreatment is conducive to the individual with BPD experiencing many current relationships as neglectful or attacking. Childhood trauma that eventuates in later BPD is viewed as associated with a concomitant unresolved and overwhelmed state of mind re-

130


garding attachment. Patients with BPD frequently present by way of violent attacks on their own bodies or those of others. The question is what additional component predisposes people with BPD to act out physically rather than in their minds. The component that is suggested is BPD individuals’ inadequacy to understand their own and others’ behavior in terms of mental states. There is also an incoherence in their self-structure that produces a split-off part of the self referred to as an “alien self.” When this alien self cannot be externalized, the experience is that of a lack of agency that is humiliating and shameful. To kill may then be an attempt to destroy a part of themselves.44 One manifestation of this inadequate understanding is an inability to represent ideas related to aggression and attachment. Affect is then used instrumentally to manipulate rather than to communicate, so that thoughts and feelings are expressed through physical action. Because borderline individuals have difficulty integrating early forms of representation of others, a quality of rigidity exists in the way they experience their inner world and that of others. As a consequence, they lack flexibility in how they represent others’ mental states. Enacting and reenacting are relied on in place of remembering. This behavior may involve provocative actions toward others that help the person with BPD experience feelings. However, this treacherous type of interpersonal relating can lead to many misinterpretations and raises the risk of violence. Self-destructive acts and suicide are always possibilities in BPD, but men with a borderline disorder are more prone to show physical violence because the independent mental existence of others poses a threat. This threat is connected to BPD individuals’ limitations in interpreting or sensing what others’ intentions are. Men with BPD may then resort to physical actions to influence or try to control others, which raises the odds for violence. In a study of 22 mentally disturbed criminal offenders who had perpetrated crimes against persons, 82% had been abused as children compared with 36% of a general psychiatric control group and 4% of normal, healthy subjects.45 Their anger toward attachment figures was described as “intense.”

EPIDEMIOLOGY OF VIOLENT BEHAVIOR IN INDIVIDUALS WITH BORDERLINE PERSONALITY DISORDER Epidemiological questions regarding individuals with BPD and violent behavior are difficult to answer because of the continued uncertainty about the etiology and pathogenesis of these individuals’ violent behaviors. The estimate is that BPD affects 2% of adults in the United


131

States, with accompanying high levels of psychiatric care and contact with social agencies.46 Intriguing questions, such as the following, have been raised: • Do BPD individuals exhibit homicidal types of violence more frequently than some other comparative diagnostic group? • Do they exhibit more homicidal types of violence compared with a control group selected from the general population? • More specifically, do they behave in a violent or dangerous manner more frequently than those with some other psychiatric diagnosis, such as schizophrenia or antisocial personality disorder? • Does the dangerousness present in individuals with BPD have any specificity connected to it in terms of which kinds of homicidal violence accompany particular aspects of their personality organization? • How does disorganization in attachment pathology contribute to homicidal violence? • Do certain childhood maltreatment situations related to BPD increase the later likelihood of homicidal violence? Answers to unresolved questions about the underlying basis for a borderline personality structure will have a bearing on the assessment of homicidal behavior and may reflect a manifestation of these individuals’ instability. Perhaps more sophisticated taxonomic devices will allow confirmation or disconfirmation of whether some of the violence seen in borderline disorders is a manifestation of an affective illness or reflects a comorbidity situation for those with this diagnosis. The tendency of individuals with BPD to misinterpret comments and behaviors and take umbrage quickly because of their oversensitivity leaves them prone to striking out both verbally and physically, often to an exaggerated degree (e.g., “to teach [those who have offended me] a lesson”). As a consequence, their interpersonal relations are often on edge. Given the chronic and episodic symptoms and traits of those with a BPD diagnosis, the question could be asked why more BPD individuals have not been involved in homicidal behavior. An initial answer is that perhaps many more borderline people are involved in serious violence than has been realized. This lack of recognition may be attributable to several factors. One reason is the absence of an official diagnosis of the disorder before 1980, although BPD was applied to diverse types of patients for decades. Hence a variety of diagnoses may have been given to people who had engaged in homicidal behavior and who would now be diagnosed as having BPD. As noted earlier, many indi-

132


viduals who become entangled in the legal system in diverse ways still do not receive an adequate diagnostic assessment, which minimizes the true incidence of violence in those who carry this diagnosis. Another reason for the difficulty in assessing to what degree BPD individuals may be involved in violence and, more specifically, the difficulty in predicting when they may become violent is that they often present as being “normal.” The mass media often refer to “inexplicable killings,” supposedly performed by a person heretofore seen as normal. Although “normal” may be used here in the sense of having no previous hospitalizations for mental illness, it suggests the possibility of a borderline personality organization operating. Casual friends and associates, based on their observations of the person’s surface-level behavior, often see individuals with BPD as socially involved, although sensitive and having some interpersonal problems—but they do not see them as misfits. In addition, the individual with BPD may be vocationally successful, although performance may be erratic. Few longitudinal studies carried out with individuals with BPD have included the variable of violence. Investigators in one study followed 62 males who had once been hospitalized for treatment at the New York Psychiatric Institute.47 Their files were later assessed, and the individuals were found to meet criteria for what we would now diagnose as BPD. There was then a retrospective follow-up over various lengths of time based on when the person had initially been hospitalized. Four of the group were found to have committed murders and seven to have committed suicide, giving a rate of 20% having been involved in violent acts with fatal consequences. The investigators did not inquire about lesser degrees of violence. Another investigation based on inpatients at the Institute of Living in Hartford, CT, found that 69% of a BPD cohort had engaged in antisocial acting out (e.g., violence toward property of others, criminal acts, promiscuity, running away, serious substance abuse problems) before hospitalization.48

CHARACTERISTICS OF BORDERLINE PERSONALITY DISORDER THAT PREDISPOSE TO VIOLENCE Many of the traits associated with the diagnosis of BPD listed in DSM-IVTR9 would seem to carry a predisposition toward violence. The group of cases used herein for illustration fulfill these diagnostic criteria, as well as criteria from the diagnostic interview for borderline patients.2 However, confining assessments of dangerousness to descriptive criteria may exclude some of the protean manifestations that add to violent propensities,


133

so that some characteristics are missed. Limitations exist from the equal weighting given the nine criteria in the diagnostic manual. However, such limitations do not foreclose discussion of the dangerousness potential inherent in these individuals’ behavioral manifestations. In the case examples used in this chapter, the individuals have been the principal participants in criminal litigation, civil litigation, or both that have been precipitated by their behavior. The intriguing questions are the tendency for violent behavior of other individuals with BPD and what cues exist to show that an individual may be on such a pathway. Two characteristics of BPD individuals’ behavior provide the possible mechanisms for a relationship between BPD and the occurrence of homicidal violence: unstable moods and impulsivity. In some persons, these moods are connected to the enduring hatred that many with this diagnosis have, whereas in others the moods are part of a concomitant depressive picture. The crucial point would be where the person’s capacity for accumulated rage intersects with his or her high potential for an impulsive outburst.

Affective Instability Depressive Symptoms The difficulties in offering specific hypotheses at this time about violent behavior in BPD individuals become evident when noting that a prevalence rate of 40%–50% for major affective disorders is found in borderline patients when using Axis I phenomenology.49 Some researchers believe the prevalence may be lower.50 However, when Zanarini and Gunderson51 used a lifetime prevalence for the diagnosis, they found that 100% of the those diagnosed with BPD met DSM-III criteria for an affective disorder. Another 84% also met criteria for substance abuse, an area that needs a similar critical assessment. Depressive symptoms may raise the risk of individuals with BPD behaving violently. All the factors that lead to violence in some depressed people can also be present in BPD individuals with a comorbid diagnosis of depression, which increases the risk of violence from the joint effect. Sometimes it is their brooding sensitivity that puts them at risk. At other times, their irritability may lead to an impaired capacity to regulate their reactions, so that impulsiveness and violent states emerge. In some cases, their provocations elicit retaliatory behaviors in others. In other cases, the behaviors of BPD individuals are a result of projective identification defenses in which they believe their own anger or rage is being directed against them by another, and they take action accordingly. The instability

134


is not confined to suicidal behaviors or gestures. The most striking quality about the depressions of patients with BPD is the accompanying instability often connected with their sensitivity to rejections. As Stone52 stated, “Minor events lead to major upsets; major events that most people take in their stride lead to catastrophe” (p. 304). Signs and symptoms of a severe depression may emerge on top of a chronic type of characterological depression. This phenomenon of “double depression,” which has been widely discussed in the literature, may also mask the potential for violent tendencies.53 On one hand, individuals with BPD have a notable lack of resiliency and adaptability. On the other hand, they have a low capacity to tolerate the pain associated with depression, further adding to their homicidal risk.54 Hypotheses about biological mechanisms operating in more minor self-injurious behaviors in BPD have also been noted and may similarly apply to the occurrence of more serious violence.55 The depressive experiences in BPD can also be seen as having specific qualities of emptiness and anger, signifying an “angry depression.”56 These depressions show significant affects of anger, anxiety, and fear, with primitive forms of object relations. The depressive experiences are characterized by chronic feelings of loneliness, emptiness, and boredom, but not guilt. When assessed by the Mutuality of Autonomy Scale, object relations of BPD individuals were seen as primitive destructive forms, with themes of torturing, strangling, and parasitic relationships or of being overpowered, devoured, or swallowed up (which tie in with masochistic themes).

Suicidal Tendencies It is interesting that clinicians are more attuned to the possibility of suicidal rather than homicidal behavior in connection with BPD. Perhaps this lack of attention to the homicidal behaviors of individuals with this diagnosis exists because following a homicidal act, these individuals are then shunted into the legal system, where they are less likely to receive clinical appraisal and focused research than are their suicidal counterparts. One long-term follow-up study of those with a BPD diagnosis indicated a suicidal risk of 5%–10%.57 The suicide rate has been found to be 55 times higher than that of the general U.S. white population.58 Continuing alcohol abuse, chaotic impulsivity, and a history of parental brutality or sexual molestation are cited as the keys to suicide in these individuals.59 Perhaps the same variables that contribute to suicidality—such as a sensitivity to rejection and reactions to the threat of loss—operate with re-


135

gard to the potential for homicidal behavior as well. For example, when BPD individuals’ close relationships are disturbed, they try to manipulate the situation to regain control; when the control does not materialize, they become more desperate, leading to feelings of abandonment and rage. A state of increasingly precarious self-regulation and incoherence may follow, with the potential for impulsive acts.

Self-Destructive Behaviors Involving Danger to Others Recurrent suicidal attempts or gestures are dangerous states and may occur multiple times in individuals with BPD. Although these acts do not necessarily involve the risk of direct violence to others, the safety of others can be impinged upon because of these individuals’ poor judgment or impulsivity. Such behavior includes driving at high speeds, various accident-prone behaviors, and suicidal gestures. In one case, a patient engaged in a histrionic ploy of leaving the car running in a closed garage while not in the car. However, this person had forgotten that a child was asleep in the house and that the gas fumes would escape into the child’s bedroom from the attached garage. The child died of asphyxiation, and the person was charged with homicide. Another area that has produced debate is whether the traits in antisocial, histrionic, narcissistic, and borderline personalities reflect separate entities or are different manifestations of one underlying personality disorder. In the background is how the patient’s capacity for displaying dramatic, self-absorbed performances might also be reflected in these other diagnoses. The features of intense, brooding anger coupled with mood instability may coalesce in a quest for self-vindication. Feeling that they have been used by others, these persons then search for an occasion to strike back self-righteously. They make rash decisions, resulting in an impulsive act, sometimes of violence, and even homicide. The act may occur in the context of feeling panicked about having been abandoned and needing to confront functioning on one’s own. Rupturing such dependency ties elicits destructive rage toward those whom they blame for their current state. Although many issues remain unresolved, this hypothesized series of events, in which these personality traits lead to violent episodes, can be illustrated in actual cases.

Cyclicity of Affective States One consequence of the affective instability in BPD is that individuals tend to cycle through emotions quickly. For example, one day their mood may be quiet and sad, with some verbalizations of hopelessness, and the next day it may be euphoric. However, relief from their depressed mood

136


seems to be connected to the intensity of their anger and hate rather than to the cycling of their emotions per se. If too much anger is abated or their anger is abated too quickly—such as may happen when it is treated with psychotherapy or pharmacotherapy—their depression may intensify. Deprived of one of their typical mechanisms for handling depressive affect—their anger—they may make impulsive decisions, such as signing out of an inpatient unit, stopping psychotherapy, ceasing medications, or making more dramatic gestures. Later they rationalize their leaving treatment, in terms of either thinking that it did not help them or feeling that the therapist was not sympathetic enough to their suffering. Such complaints are similar to those that occur in the context of BPD individuals’ relationships with intimates or in their daily social lives. Manipulative behavior is a common modus operandi for coping with minor affective disturbances. What is often missed is the degree of BPD individuals’ anger with the potential on some occasion to go beyond their “affective storms.” A succession of people in therapeutic roles, from a variety of mental health fields with different treatment modalities, may have fused the BPD patients’ past and present hatred and anger toward others into now focusing on one person or group. However, the everready potential to act out an angry role as a miscarried attempt at selfhelp remains. Individuals with BPD, like those with antisocial personality disorder, may begin to “burn out” over time, and on long-range follow-up they have different personality disorders such as histrionic or avoidant.60 However, that may not occur with respect to the younger subgroup, who verge in and out of the “border” of dangerous behavior and may come from more traumatic developmental backgrounds. A typical situation arises when a seemingly unpredictable outburst of high-risk behavior occurs. The behaviors may fluctuate with such rapidity that a clinician (let alone others who are in contact with the patient) cannot keep ahead of the patient. It is not unusual for an individual with BPD to have engaged in a pattern of secret and devious behavior over an extended period of time that is suddenly exposed and causes the person great shame. The homicidal message is sometimes conveyed within a family. In one case, a serious suicide attempt by a 14year-old boy revealed a history of his mother frequently threatening to commit suicide because of her unhappiness and because he and his father did not meet her needs. Often in retrospect, when an in-depth clinical history is put together, it is easy to see how subtle patterns of eccentric behaviors have operated in a person and have been hidden for some time. Many times, the history will reveal that that person behaved in unpredictable and unstable ways for a while. In a statistical sense, because dangerous behavior


137

represents only a small number of the deviant behaviors engaged in, predicting these rare events will always be a difficult problem. Apart from the obvious—that instability in any setting increases the likelihood of a violent act—the question is whether any specificity is connected to instability in individuals with BPD. Excessive adoration and groping for someone to idealize are doomed to result in disappointment given human nature; and in BPD individuals, this disappointment progresses to a state of felt betrayal or a need for revenge in the context of unremitting hatred. In the midst of a divorce from his wife with BPD, the husband stated, “I could handle all the upsets, lies, and uncertainties, but I couldn’t continue to live with someone I felt hated me so much.” The BPD wife was later charged with conspiracy to commit murder when she tried to hire someone to kill her former husband.

Impulsiveness When occurring together, two of the BPD criteria listed in DSM-IV-TR carry a high potential for dangerousness: 1) unstable and intense interpersonal relationships and 2) impulsiveness in potentially self-damaging areas. Background subtleties illustrate how potential violence is reached. In the following case example, impulsiveness is viewed as a personality trait and does not indicate merely an occasional impulsive act.

Case Example A woman in her late 20s was a defendant in a criminal suit as well as a civil lawsuit brought by a female colleague in her former company and their joint male supervisor. Initially, the woman had seen herself as the favorite of the supervisor, an assessment made on the basis of his bantering with her when alone or when mingling with a group of fellow employees. Although her interpretation was that she was being singled out for this special treatment, other employees described the supervisor as a generally gregarious and friendly person. The situation became a crisis when the woman realized a new female employee had begun dating the supervisor. She then shifted into devaluing him as “dishonest, worthless, and a person who used people.” Previously a good worker who “performed for him,” she began to brood about the “betrayal.” While at dinner with a friend, she found herself talking “out of control about that no-good bastard.” Although not a drinker, she became intoxicated and went back to the office at about 10:00 P.M., thinking that he might be there and hoping to have a confrontation. Finding him absent, she proceeded to wreak havoc on several offices, including her own. Special effort was taken to destroy personal effects and mementos in the offices of the supervisor and the employee he was dating, along with slashing furniture with a knife.

138


While leaving the scene in her vehicle, she was driving at a high speed and collided with another vehicle, in which a person was killed. Although nothing had been reported stolen from the offices, the vandalism indicated another basis for a charge. Thousands of dollars in damage had been done, and criminal and civil lawsuits were initiated. On psychiatric evaluation, the woman expressed gratification that the supervisor had not been present because she had had the conscious thought of killing him that evening, possibly killing herself, or both. Instead, an innocent driver was killed in the course of a series of impulsive acts. An interesting parallel existed from the woman’s college years when she ran for a sorority office but lost the election to a friend. Her view of the friend abruptly shifted to one of someone who had betrayed her by entering the race after she had. A few weeks later, she misinterpreted a comment made by the former friend at a dinner meeting and shoved a piece of pie in her face.

This woman had a pattern of chronically unstable and intense interpersonal relationships that were ordinarily concealed by her dedication to hard work, her physical attractiveness, and her ability to pick up new relationships and abandon old ones. Although such relationships rarely moved beyond a superficial level, they allowed her to adapt superficially while not revealing her hidden vulnerabilities. The impulsivity of her behavior on the night in question was not the first of such behaviors, although other events had usually been concealed and had not involved homicidal intent. Sometimes the patient’s history reveals a heightened propensity to react as described here in response to misguided advice from a friend or professional person. This other individual may be manipulated into condoning or acquiescing some initial steps on the basis of the patient’s pleading behavior about how he or she has been wronged. In some cases, therapists with unresolved problems of their own in this area may promote the patient’s behavior as a way of obtaining a vicarious revenge from their own unresolved conflicts. The rash advice offered may be a nonviolent course of action, such as separating from a marriage, breaking up a dating relationship, filing a lawsuit, quitting a job, moving to another part of the country, and so on. When the patient’s decision involves an overt act of responding to an affront—real or exaggerated—and a felt need for revenge, the possibility of some degree of harm to another, including homicide, arises.

Case Example A case that began in a seemingly innocuous manner quickly escalated. A divorced woman bought a new house. When her former husband learned about it, he brooded about how unjust it was that her life was


139

now much better than his. Over the years, he had been treated for depressions combined with impulsive behavior. One night he ventilated about the situation in a group therapy session. The therapist had the participants vote as to whether his anger was justified, and his fellow therapy patients voted that it was. Later that night, during a rainstorm, the patient drove onto the front lawn at his former wife’s new house, leaving deep tire ruts that later required a new lawn to be laid. Police were called and a high-speed chase ensued, with shots exchanged between the police and the patient. The patient was wounded and later charged with attempted murder. When seen for an independent psychiatric examination for legal purposes 1 year later, he was still clinging to his belief that his behavior had been justified, based on the vote in group therapy, and still ruminated about ways to hurt his ex-wife.

Rage, Hatred, and Tantrums Separating the brooding anger displayed by individuals with BPD from a periodic paranoid ideation may be quite difficult. Rage reactions draw on anger as well as paranoid thinking and, once again, the problem of comorbidity arises. There are degrees of aggressive affect—irritation, anger, hatred, and rage. Rage is an acute reaction, whereas hatred is a more chronic state that in its typical form seeks to devalue an object and in its extreme form seeks to eliminate it symbolically or directly.

Case Example After her marriage ended in divorce, a 45-year-old woman began to trail her ex-husband in a detective-like fashion and amass data on him. This occurred without an expressed desire on her part to reconstitute their marriage. In earlier marriage counseling, their relationship had been described as having unrequited affection and being better ended. However, after it did end, the wife brooded about revenge to demonstrate what a “womanizer” he was. Although she could discover only one woman in his new life, she became convinced that he had relationships with several women. Her anger at his “deceit” was intense and unremitting, and it did not wax and wane as it had during her earlier depressive states. In time, stalking her ex-husband and his lady friend to restaurants or to their respective apartments did not suffice to calm her. As a next step, she obtained the woman’s name, called her, and introduced herself as the ex-wife. She spoke in a pleasant voice to the woman on the phone and arranged a luncheon with the woman, throughout which her pleasant attitude persisted. After the luncheon, the new lady friend became suspicious of the ex-wife because she saw no purpose in pursuing the relationship. When the husband’s lady friend refused subsequent offers to meet after the initial luncheon, the ex-wife’s anger grew. The ex-wife then made a series of phone calls to the woman, telling her, as the ex-wife put it, “all the bad things I could think of.” Even the

140


woman’s decision to hang up the phone without listening each time the ex-wife called, in the hope that the calls would cease, had little effect. Eventually, the woman called the police to report the harassment. At that point, the ex-wife decided to turn the tables by reporting her former husband and lady friend as the ones who were harassing her by making nuisance calls with lewd comments. This stalemate then shifted to a step-up in dramatic actions. The ex-wife went to the president of her former husband’s company, whom she knew as a person of strict morality, and told him that during the last years of their marriage, her husband had been having affairs. This led to the president calling the husband in; the husband denied the ex-wife’s allegations and became enraged about her lying. In response, he went to his ex-wife’s residence, where a shouting match led to objects being thrown, physical pushing, his physically assaulting her, and, eventually, the ex-wife stabbing and killing her ex-husband in what she described as self-defense.

This case example illustrates the desperate efforts of a person with BPD to deal with the recurring depression and rage she felt subsequent to a loss and how the mishandling of the loss resulted in a homicide. Her feelings of helplessness led to anger and rage as a response to her injured self. If she could somehow punish her husband, the hope was that she could regain some sense of power and equilibrium. Murder became the last resort to what seems like a state of unrelieved victimization. The ex-wife’s increased risk-taking behaviors and tantrums reflected her attempt to relieve her paranoid brooding and regain control by going on the offensive. Her stalking behavior was similar to that seen in others with BPD whose impulsivity becomes more prominent as their control over it falters. Those who stalk celebrities or political figures often seem to have a background of brooding anger and mood instability, characteristics that carry the potential of danger. Cases in which an individual has depression mixed with rage reactions have the potential for tragic consequences, such as the “fatal attraction syndrome.” The movie Fatal Attraction has been critiqued from a feminist viewpoint as portraying a certain type of woman as the provocateur in this type of situation.61 The following clinical case antedated the movie and again illustrates the stalking theme.

Case Example A young female in her 20s dated a man for more than a year, after which he broke off the relationship. He subsequently began to date another young woman, whom he married 9 months later. A few weeks after returning from their honeymoon, the new husband returned to their apartment to find his wife lying dead on the floor. An autopsy listed 97 stab


141

wounds on her body. After first being a suspect, the husband was cleared and the investigation eventually led to the former girlfriend who, it was determined, had earlier been stalking the couple in their apartment. The girlfriend underwent hypnosis and amobarbital interviews, arranged by her legal defense team, through which a fragmented story emerged. Based on that story, an insanity defense was entered. Throughout the pretrial and trial, the woman presented the image of a buoyant, talkative, flirtatious, and denying young woman who seemed dissociated not only from her act of homicide but from the entire proceedings. The woman was eventually found not guilty by reason of insanity. The drama did not end there; a subsequent legal case arose out of her desire to marry a fellow inmate while she was confined in a state hospital as mentally ill and dangerous. Although her reality testing seemed intact, her feeling of reality was impaired by her dissociative qualities.

Emptiness, Loneliness, and Fears of Abandonment Accompanying many of the illustrated behaviors are states of desperation experienced as feelings of emptiness and aloneness. Defensive activities, such as frenzied social and work overcommitments to avoid being alone, are well-known patterns that individuals with BPD follow to temporize the loneliness. Even for the group of BPD individuals with adequate social skills who mingle with others, there is loneliness and a sensed lack of fulfillment in their relationships. What persists is a quest to blame others for their unhappiness. When disappointments occur in the course of life, their rage accumulates, along with the potential for someone to become the recipient of their stored rage in an actual or symbolically murderous act. The competence of BPD individuals is only an “apparent competence” given their heavy reliance on others. Such reliance makes the possibility of abandonment more threatening.62 To head off their abandonment fears, these people engage in a variety of threats, manipulations, claimed entitlements, blackmail, and maneuvers, including using children to attain their ends. It is when these measures are unsuccessful that the individual escalates his or her behavior into more high-risk behaviors that involve the possibility of violence. The following is a more extreme example of violence occurring in connection with BPD. This example involves a micropsychotic episode that evolved with paranoid thinking.

Case Example A recently separated woman who had been treated for depression by various therapists in the past became anxious about living alone and feeling a lack of support. She began to wonder whether her phones were bugged

142


or if others had obtained a key to her house from her former husband. Although not sure about these ideas, she changed the locks and initially felt reassured when her young daughter was with her. However, the feeling of security did not continue and she began moving from one hotel to another or staying with relatives or friends. When she had exhausted all of these alternatives, a series of panic attacks occurred. Some of these involved intense fears that she was being followed when driving. Matters were brought to a head one evening when, in an attempt to lose her supposed pursuers, she began speeding on a freeway to escape cars she classified as either “friendly” or “unfriendly.” Feeling her anxiety mounting, she decided to get off the freeway and drive to the first office building she could find. Exiting the freeway, she feared it was too dangerous to park in a lot and instead drove her car directly through the front doors of an office building. She then sat in the locked car until she felt reassured that the people surrounding the car would not harm her. Unfortunately, someone who had been near the doors she had driven through later died from injuries sustained through the car’s entrance.

Substance abuse problems are another complication in BPD individuals’ effort to combat loneliness. However, substance abuse also raises the potential for violence. For those who are not self-supporting, chemical dependency may lead to illegal behaviors, such as involvement in prostitution or selling drugs. Others begin the use of such substances during work hours, thus jeopardizing successful past employment; threats to their job performance or security then affect their vulnerable status. Some use drugs to satisfy their need for sensation seeking or to engage in high-risk activities. Whatever the basis, substance abuse heightens the possibility of violent activities in people with BPD. The following case example illustrates the destructive combination of abandonment fears with substance use.

Case Example A well-groomed and stylishly dressed 32-year-old man sought help for recurrent depressed states. He had undergone several earlier attempts to treat his depression, varying from psychoanalysis to pharmacotherapy. Although he had always had a succession of women in his life, he needed constant reassurances that they would remain loyal to him. Eventually, he would begin nagging them about their activities and possible faithlessness to him. The tempo of these verbal exchanges would increase, and the confrontations between the man and his girlfriend would become accusatory. The man’s spying on these women in their places of employment, casually dropping by too often, or standing in nearby office buildings to see whom they left with at lunch or at the end of the day would eventually precipitate crises. The outcome for the man was a rage reaction; he would explode with accusations and often be-


143

come physically abusive. In some cases his behavior would end the relationship, whereas in others it led to efforts at being reunited, seeking counseling for codependency, and other helpful interventions. In the context of breaking up a relationship, or before a new relationship began, this man would be assailed by feelings of being alone in the world with no one close to him and then begin to visit prostitutes. Action-oriented behaviors, such as high-speed drives at 100 miles per hour, were another part of his pattern. These behaviors might involve drag racing against anonymous others. Some of these ventures would be carried out under the influence of drugs or alcohol, and the man would eventually end up “thawing out.” He once entered treatment for chemical dependency and was told by a counselor that he was addicted to one of the medications a previous psychiatrist had given him but now refused to prescribe by phone. The man continued to obtain the drug by visiting various physicians’ offices. This man eventually became entangled with the law on several occasions, such as being charged with attempted murder when he got into a fight that resulted in the victim lapsing into a coma for days with a skull fracture and subdural hematoma, but surviving. There also was a civil lawsuit for harassment resulting from his appearing at a woman’s place of employment so frequently that his behavior was noticed by fellow employees and her supervisor. On another occasion, he filed an ethics complaint against the psychiatrist he felt had made him dependent on the medication and whom he later blamed for his difficulties.

Endless legal involvements often loom in these types of situations, based on events such as hit-and-run accidents, fights or quarrels with people whom the individual with BPD believes have slighted him or her, and assaults on friends or lovers.

Disturbed Sense of Reality Another characteristic that may emerge in those with a borderline personality organization is a disturbance in their sense or feeling of reality. This disturbance is different than their reality testing being impaired, such as in a psychotic organization, where differentiation from others is difficult. Individuals with BPD are usually capable of distinguishing their inner life from the world of external events, apart from during any micropsychotic episodes that may be experienced. However, their confused social reality creates difficulties, especially a difficulty in discerning how and why people and situations may be viewed differently on different occasions. Such a cognitive fixity gives a rigidity and an impaired capacity to read others and to respond, which is a deficit in empathy. It leads people and organizations with whom they deal to misjudge their actual social capabilities based on external behaviors. What goes for social poise may be a thin veneer for insecurities and lurking misinterpretations.

144


Splitting as a Defense To further understand the proneness to violence in individuals with BPD, the theoretical construct of splitting as a defense is often introduced. It is well known that these individuals oscillate in their assessments of others between idealizing the others as good and devaluing them. The shift to badness raises the potential for their seeing another as a persecutor. Diverse primitive defenses related to the splitting are projective identification, omnipotence, omnipotent control, primitive idealization, and devaluation and denial.63 These defenses induce an unevenness in personal relationships that contributes to impulsive behavior. The individuals shift from manipulativeness and arrogance to helplessness and seeking to blame others for their predicaments. In the midst of poorly regulated anxiety, depressive states and rage emerge, with the possibility of acting on the murderous rage. Superego functioning also makes its contribution to the imbalance between the punitive aspects of some wrongful act or fantasy of an individual and the idealized notion of how he or she should perform. The result is the lack of a stable, internalized system of morality.64 On one hand, guilt and depression emerge; on the other, there is manipulation and exploitation. What this duality reflects is splitting into the extreme of the all-good or all-bad representations of the self and others. Rather than being realistic ideals, the ideals are driven by power, domination, and perfectionism. One reason for the lack of integration is the intensity of aggression influencing self and object representations. Attempts to bring together the extreme degrees of love and hate toward the self and others initiate confusion, anxiety, and depression. The theory is that conflict over aggression in childhood during the oral and anal stages is later projected, leaving the person prone to attribute a paranoid imagery to parental figures later in life, who are then seen as dangerous. These confused internal images later leave the person with BPD vulnerable when disagreements arise and thus tend to make him or her more likely to act on his or her hatred.

Disconnected Quality of Behavior Another contributing variable to violent behavior—smoldering resentments and enduring hatred—may seem incongruent with the theory that individuals with BPD frequently vacillate between opposites. In line with this characteristic, these individuals often do not seek just a vindication or victory but sometimes a total annihilation of a selected target who becomes a hated object and once again is seen as responsible for their unhappy plight. Such targets may be chosen from anyone and


145

anything connected to an event or time in the individual’s past or present life. Although tantrums and rage reactions occur, such “ventilating” at best only leads to a temporary alleviation of the individual’s turmoil. As one individual put matters after an attempted murder had gone awry, “I may forgive but I never forget.” Distorted memories persist and contribute to future blowups as fantasy stirs up needs for future reenactments. If there is no current provocation, past ones will serve. Perhaps some past wrongs to justify their acts need to be embellished or even invented. Such confusion may help explain the pseudologia fantastica seen in persons with BPD.65 What results is a disconnectedness in behavior from long-term memory. Assessments about probabilities of behavior are decoupled from the overall assessment functions that customarily operate in people. Unmodulated responses occur abruptly and may not be in direct response to a present situation. An unpredictable and exaggerated response, to the point of violence, corresponds to this type of instability. Hence the minor slights or rejections become monumental. Competition that emerges in some areas of a BPD individual’s life (e.g., those involving appearance, work, social life) is perceived not just as competition but a threat to his or her life, prompting the individual to rise to the level of needing to destroy someone. Such an overlearned memory maintains old grudges and misinterprets them in their present environment. The “sore spots” persist and later trigger stereotyped maladaptive responses. Maltreated or abused children or adolescents may not only overreact to those in authority or those of the opposite sex but, as they develop, may also remain prone to misinterpret overtures, gestures, or words. They remain ripe for making accusations about another, taking umbrage, or perhaps attacking others outright—if not physically, at least verbally or in some devious way. In a longitudinal follow-up of 206 former inpatients with BPD, an analysis of variance found parental brutality as the factor most likely to predict a worse-than-average outcome.66 Poor outcomes were noted in those who had been jailed following behaviors based on impulsivity or flaunting the law or who had developed counteraggressivity as a result of parental cruelty. Clinical descriptions of dissociated states with dramatic alterations in personality have been present since the nineteenth century. Such splitting as part of a borderline syndrome helps explain how some homicidal-level acts can occur. The hallmark is the fluctuating emotional state of patients with BPD, with their accumulated and slowly subsiding rage. Situations go awry if an attempt to dissipate their hate is too rapid, such as may happen in the course of some treatment processes,

146


or in therapeutic relationships that tap their profound ambivalence. Disruptions of attachments in daily life provide another explosive possibility. When a person with BPD experiences such a lack of homeostasis, a dissociative episode may intervene that allows him or her to flee into an altered personality who then commits a violent act. In an altered state, diverse types of destruction may occur: suicide attempts or gestures, head banging, wrist slashing, object throwing, attacking of others, or a homicidal-level attack. Murders that seem particularly brutal often seem mystifying, especially in the overkill of a loved one. For example, why stab someone repeatedly when a few times would suffice, or why mutilate a body? In effect, this is like killing the victim multiple times. Later the person often does not recall anything close to the extent of the homicidal acts. One explanation for these complex murders is to connect them with having been shamed and humiliated, so that when a homicide results, the perpetrator is in a state of annihilating many times over the victim who has induced the shame. An abrupt shift into a blatant, smiling, la belle indifférence may subsequently appear. BPD persons may manifest “attacks” at sites connected with the deceased, such as funerals or ceremonies, in which their external signs of grief may be conspicuously striking. One perpetrator of a homicide, undiscovered at the time, joined mourners in expressing his deepest regrets about the death of a loved one. In one case, a person later indicted for a homicide attended the victim’s funeral. Mourners were impressed by the depth of grief they witnessed in him. An interesting twist in this type of situation occurred when a woman reported to the chair of a company selection committee that when she was a coworker, she had once carried on an affair with the leading candidate for a senior position. Sometime later after the man had committed suicide, the woman sent condolences to his wife, along with offers of friendship, although the woman and the wife had never met. Such vicarious killings by way of “killing his chances” are rarely uncovered publicly.

LEGAL CONSIDERATIONS: PERSONAL RESPONSIBILITY AND BORDERLINE PERSONALITY DISORDER Given the traits and behaviors of individuals with BPD, a recurring problem that arises often in legal or ethical contexts is the question of their responsibility for their behavior. This question is more challenging for this group than for those with any other diagnosis. Although a psychiatrist becomes aware of the dynamics and interactional systems that reveal shifting patterns between the victim and the victimizer or the op-


147

pressed and the oppressor, issues involving appraisals of responsibility do not usually arise in clinics, and addressing such questions is usually discouraged. However, such bypassing does not occur when the behavior of BPD individuals leads to legal charges. Responsibility questions are not confined to the criminal arena, because civil actions may be later instituted against perpetrators of homicides for wrongful death actions. Alternatively, the perpetrators and the attorneys hired to represent them may raise issues about the limitations of perpetrators’ responsibility for their behavior, Often persons with BPD see themselves as being the wronged party. Also, these patients’ life orientations may be litigious ones in which they have initiated suits. Problems arise when they become the aggressor because of their propensity to react or to exact revenge. In some cases, their previous physicians, lawyers, or therapists become potential victims, sometimes on a homicidal level, when BPD individuals believe they have been abandoned. The end of a friendship can similarly produce a feeling of betrayal and powerlessness and lead to homicidal behavior. In the criminal area, the issue of responsibility may arise through some variation of an insanity defense, diminished-responsibility approaches, or allegations that sexual or physical abuse is responsible for the behavior. The area is a quagmire of confused efforts to handle legal issues, given the lack of conceptual clarity in psychiatric thinking about which types of personalities are to be seen as responsible for homicidal behavior. Some argue that such questions should be dealt with outside the province of medicine or psychiatry. However, it is difficult to bypass the issues because, at least implicitly, the questions touch on psychiatric preconceptions about behavior on many levels. A key issue is how and to what degree BPD individuals’ capacities to exercise choice and to make decisions are affected by their personality functioning. Analogous quandaries arise in the frequently encountered area of borderline individuals who also have alcoholism problems and who become involved in legal difficulties associated with their propensity to become intoxicated and act in destructive ways. Although the emphasis in treatment programs dealing with alcoholism is that such people have a disease and may or may not be in control of themselves, exercising choices such as entering treatment programs and giving up drinking is implicit in the programs’ approaches. Some programs are particularly critical of the viewpoint that alcoholic individuals lack free will and should not be viewed as responsible for their behavior.67 Similar questions arise with regard to other comorbid disorders that may be present in individuals with BPD. Allegations of behavior occurring in trancelike states, fits, or fugues or through multiple personalities

148


may be raised. For example, recently efforts have been made to extend the boundaries of nonresponsibility to include violent acts ascribed to premenstrual syndrome, battered woman syndrome, or battered child syndrome, if they are seen as part of the picture of borderline personalities. The reasoning is that the perpetrator has developed such a physiological or psychological vulnerability, or has been living in such a state of fear from past assaultive behavior, that he or she believes his or her life is in danger. Hence, the person lacks the capacity to make a voluntary choice and interpret a situation as either “the other person’s life or mine.” A similar position has been argued in the case of people with BPD who endured physical or sexual abuse as children and later committed violent acts. The conclusion urged may be that the person should not be seen as currently responsible for his or her actions because the impact of past traumatic occurrences has impaired his or her judgment. The problem is that such a position is unsupported by empirical data about the long-term consequences of abusive home environments in connection with adult violence.68 Not all abused children emerge with negative psychiatric consequences from abuse, let alone homicidal potential.69 The clinical question is whether the specific adverse consequences can be demonstrated for a particular child. These exculpatory issues are troublesome because of the increased frequency with which they are raised in connection with the BPD diagnosis. The question is the degree to which such behavior reflects some type of cognitive dysfunction, even given that the choice to commit violence is a troubled one. Can these individuals control their behavior to a degree similar to others who have certain personality traits and experience life stresses? Why does a particular BPD person commit a homicide whereas others with the same diagnosis do not? Specifically, could the BPD person have done otherwise but simply chose not to? The clinical inferences of this situation can be examined without becoming lost in an endless philosophical debate as to whether someone who ought to behave in a certain way can do so. It is logical to distinguish the overall group of BPD patients from the subgroup of those individuals with this diagnosis who have had a major psychotic or micropsychotic episode imposed on their borderline state at the time of homicide. Even these psychotic cases raise issues related to the perpetrator’s responsibility that are not easily resolvable and are often contested in court. However, there are clinically distinguishable signs and symptoms when a BPD individual has experienced a psychotic episode. If it were to be argued that all impulsive behavior occurring in individuals with BPD occurs on a psychotic basis, which is an unproven assumption, it would shift the entire level of discussion away from just


149

BPD individuals to all psychotic people. This raises broader issues such as whether most BPD patients, if not all those with personality disorders, are to be viewed as psychotic whenever individuals with that diagnosis behave in a violent manner. Such a circular argument is likely to confuse matters further. It would not be clinically sound to classify as psychotic every eruptive and violent act by an individual with BPD. The person who destroyed the offices of her colleagues was quite willing to see herself as lacking control over her behavior. People who act from motives of spite, malice, hatred, envy, revenge, and so on are subsequently quite eager to see their behaviors as being explained, and therefore excused, because certain emotions were predominant or seemed overwhelming at the time. Assessments become more difficult when people with BPD appear to be experiencing micropsychotic episodes. For example, a complicated assessment would occur with the person who believes he or she was in danger—such as thinking he or she was being pursued on a highway, which led to his or her efforts to escape, even at the cost of driving a car into the entrance of a building and killing someone. In such cases, a psychotic diagnosis might be added. Yet in most cases, the thread of commonality in the behavior seems to be unmet power needs that are masked and rationalized by an experienced state of helplessness and lack of control. An ultimate issue operating in such cases is whether impulsive behavior occurring in a borderline personality should be seen as reflecting an incapacity to exercise choice. Or does such behavior rather reflect a person who wishes and chooses to behave in that manner from other motives? Is a person with BPD so incapacitated that he or she could act in no other way at the time of the violent act? If so, the capacity to choose was impaired to such a degree that self-restraint was an impossibility. Such a line of reasoning should presumably hold not only for a particular act, such as homicide, but also for other diverse and scheming behaviors. To argue otherwise would imply that a qualitative difference exists in such personalities that these individuals are held responsible only for some impulsive acts but not others, and more confusion would reign from trying to decipher which is which. Simon70 noted that various cult leaders display the characteristics of the borderline personality. Many of the borderline traits become exacerbated in the leader when some cataclysmic psychological stress is imminent and the cult itself may be near an end-point. The result may then be mass acts of murder and suicide. Examples are Jim Jones and his socialist utopia in Guyana, where Congressman Leo Ryan was killed and 900 members committed suicide; Charles Manson, who carried out murders via

150


his gang to try to provoke a race war that would end in “Helter Skelter,” as he put it; and David Koresh and the Branch Davidians in Waco, TX, where 82 members either suicided or were murdered or killed. The question of personal responsibility should be extended beyond the legal setting. If confined to legal situations, such reasoning would lead to a position that a valid argument for nonresponsibility can be made only when an impulsive or scheming act occurs that leads to a charge in a courtroom. The responsibility of individuals with BPD should also be questioned in connection with other types of provocative behaviors such as, for example, when a BPD individual feels a powerful desire to “do someone in” by verbal gossip or slander rather than to restrain himself or herself. Are borderline persons unwittingly in such a predetermined ego state that in one of their emotional storms, their outburst cannot be controlled any more than a homicidal act? Similar questions operate for minor outbursts such as throwing objects at people; shoving a piece of pie in someone’s face; or the little acts of lying, deception, and exaggeration that occur as part of these individuals’ daily functioning. Conversely, if each of these types of behavior is seen as one of a series of choices the person is making, with diverse motives and gratifications operating, the consequences accruing from the behavior would be interpreted quite differently. The behavior is then less readily tolerated or acceptable. Instead, it may be seen as typical of certain people by those who know them well and who learn to tread carefully around them. Such adaptation by associates should not be confused with exculpation based on the view that the person could not control his or her actions. Serious consequences may follow, depending on which view of borderline behavior is taken. If individuals with BPD are seen as out of control, what follows is a perspective that they are “sick” and not to be assessed as blameworthy. Perhaps, like other sick people, they deserve pity. Such an assessment of illness may allow a host of fringe benefits to accrue to them, such as referrals for treatment, disability payments, participation in support groups, and collection of awards in civil litigation actions—as well as permission to continue acting periodically in the manner they have previously, because their uncontrollable illness may reassert itself in the future. A contrary view takes a position that individuals with BPD are responsible for their actions. From this position, they are no longer seen as having permission to behave in the ways they have in the past or to be classified as nonculpable. The “sick” role of attributing their behavior to past adverse influences, either from genetic endowment, childhood experiences, or unfortunate encounters in adulthood, is not seen as an accept-


151

able excuse for their actions. These people would then be expected to assess their future behavior by calculating their anticipated pleasure against the painful consequences of continuing to play the odds of living in their flamboyant and provocative style. Morse71 argued that the sine qua non for assessing legal responsibility is the capacity of the individual to rationally comprehend the most morally relevant facts bearing on his or her culpability. The general theory of excuse is then based on a lack of a full capacity for rationality rather than the diagnosis itself. Barring some accompanying micropsychotic disturbance in thinking, most borderline individuals would not meet a test for lack of responsibility. Clinicians are aware of the need some patients have to increase the tempo of their behaviors when they are not being successful in achieving sought-after goals. For individuals with BPD, the resultant behaviors are tantrums, rages, and attacks. Halleck 72 commented that it is important not to assess responsibility in terms of whether the behavioral outcome seems rational or irrational. The difficulty is that treating professionals often make moral judgments and then seek to buttress their moral conclusions clinically by using the BPD diagnosis as the basis for saying someone lacked the capacity to control homicidal acts. At a minimum, a conceptual framework is needed when responsibility is tied in with a clinical goal of helping someone function differently in the future. Although this framework would not exculpate the BPD person from errant behavior in a legal sense, it would at least keep psychiatrists from becoming trapped in assessments, rightly or wrongly, in which patients and attorneys maneuver them into playing the role of moralist without the psychiatrists’ realizing it.

REFERENCES 1. Knight R: Borderline states. Bull Menninger Clin 17:1–12, 1953 2. Gunderson JG, Kolb JE, Austin V: The Diagnostic Interview for Borderline Patients. Am J Psychiatry 138:896–903, 1981 3. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC, American Psychiatric Association, 1980 4. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987 5. Clarkin JF, Kernberg OF: Developmental factors in borderline personality disorder and borderline personality organization, in Borderline Personality Disorder. Edited by Paris J. Washington, DC, American Psychiatric Press, 1993, pp 161–184

152


6. Berelowitz M, Tarnopolsky A: The validity of borderline personality disorder: an updated review of recent research, in Personality Disorder Reviewed. Edited by Tyrer P, Stein G. London, England, Gaskel, 1993, pp 90–112 7. Kohut H: The Restoration of the Self. New York, International Universities Press, 1977 8. Kernberg OF: Borderline personality organization. J Am Psychoanal Assoc 15:641–685, 1967 9. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 10. Pope GH, Jonas JM, Hudson JI, et al: The validity of DSM-III borderline personality disorder. Arch Gen Psychiatry 40:23–30, 1983 11. Stone MH: Determining amenability to psychotherapy in patients with borderline disorders. Psychiatr Ann 34:437–444, 2004 12. Kaplan HI, Sadock BJ: Personality disorders, in Synopsis of Psychiatry/ Behavioral Sciences/Clinical Psychiatry. Baltimore, MD, Williams & Wilkins, 1991, pp 525–542 13. Kavoussi RJ, Coccaro EF, Klar HM, et al: Structured interviews for borderline personality disorder. Am J Psychiatry 174:1522–1525, 1990 14. Grinker RR, Werble B, Dyre RC: The Borderline Syndrome. New York, Basic Books, 1968 15. Liebowitz MR, Klein DF: Hysteroid dysphoria. Psychiatr Clin North Am 2:555–575, 1979 16. Olsen J: Cold Kill. New York, Atheneum, 1987 17. Torgersen S, Lygren S, Oien PA, et al: A twin study of personality disorders. Compr Psychiatry 41:416–425, 2000 18. Oquendo MA, Mann JJ: The biology of impulsivity and suicidality. Psychiatr Clin North Am 23:11–25, 2000 19. Gurvits IG, Koenigsberg HW, Siever LJ: Neurotransmitter dysfunction in patients with borderline personality disorder. Psychiatr Clin North Am 23:27–40, 2000 20. O’Leary KM: Borderline personality disorder: neuropsychological testing results. Psychiatr Clin North Am 23:41–60, 2000 21. Brown GL, Ebert MH, Goyer PF, et al: Aggression, suicide, and serotonin: relationship to CSF amine metabolites. Am J Psychiatry 139:741–746, 1982 22. Siever L, Trestman RL: The serotonin system and aggressive personality disorder. Int Clin Psychopharmacol 8 (suppl 2):33–39, 1993 23. Stone MH: Towards a psychobiological theory of borderline personality disorder. Dissociation 1:2–15, 1988 24. Paris J: Antisocial and borderline personality disorders: two separate diagnoses or two aspects of the same psychopathology? Compr Psychiatry 38:237–242, 1997 25. Zanarini MC, Williams AA, Lewis RE, et al: Reported childhood experiences associated with the development of borderline personality disorder. Am J Psychiatry 154:1101–1106, 1997


153

26. Greenman DA, Gunderson JG, Dane M, et al: An examination of the borderline diagnosis in children. Am J Psychiatry 143:998–1003, 1986 27. Paris J: Personality Disorders Over Time: Precursors, Course, and Outcome. Washington, DC, American Psychiatric Publishing, 2003 28. Lofgren DP, Bemporad J, King J, et al: A prospective follow-up study of socalled borderline children. Am J Psychiatry 148:1541–1545, 1991 29. Links PS: Family environment and family psychopathology in the etiology of borderline personality disorder, in Borderline Personality Disorder: Clinical and Empirical Perspectives. Edited by Clarkin JF, Marziali E, Monroe-Blum H. New York, Guilford, 1992, pp 45–66 30. Weiss M, Zelkowitz P, Feldman R, et al: Psychopathology in offspring of mothers with borderline personality disorder. Can J Psychiatry 41:285–290, 1996 31. Bezirganian S, Cohen P, Brook JS: The impact of mother-child interaction on the development of borderline personality disorder. Am J Psychiatry 150: 1836–1842, 1993 32. Kernberg OF: Borderline Conditions and Pathological Narcissism. New York, Jason Aronson, 1975 33. Adler G: How useful is the borderline concept? Psychoanalytic Inquiry 8:353–372, 1988 34. Masterson J: Treatment of the Borderline Adolescent: A Developmental Approach. New York, Wiley, 1972 35. Abend SM, Porder MS, Willick MS: Borderline Patients: Psychoanalytic Perspectives. New York, International Universities Press, 1983 36. Hurt SW, Clarkin JF: Borderline personality disorder: prototypic typology and the development of treatment manuals. Psychiatr Ann 20:13–18, 1990 37. Fahy TA: The diagnosis of multiple personality disorders. Br J Psychiatry 153:597–606, 1988 38. Putnam FW: Diagnosis and Treatment of Multiple Personality Disorder. New York, Guilford, 1989 39. Bowlby J: Attachment and Loss, Vol 1: Attachment. New York, Basic Books, 1969 40. Main M, Solomon J: Proceedings for identifying infants as disorganized/ disoriented during the Ainsworth Strange Situation, in Attachment During the Preschool Years: Theory, Research and Intervention. Edited by Greenberg M, Cicchetti D, Cummings EM. Chicago, IL, University of Chicago Press, 1990, pp 121–160 41. Fonagy P: Attachment Theory and Psychoanalysis. New York, Other Press, 2001 42. Fonagy P, Gyorgy G, Jurist EL, et al: Affect Regulation, Mentalization, and the Development of the Self. New York, Other Press, 2002 43. Judd PH, McGlashan TH: A Developmental Model of Borderline Personality Disorder. Washington, DC, American Psychiatric Publishing, 2003

154


44. Fonagy P, Bateman AW: Attachment theory and mentalization-oriented model of borderline personality disorder, in The American Psychiatric Publishing Textbook of Personality Disorders. Edited by Oldham JM, Skodol AE, Bender DS. Washington, DC, American Psychiatric Publishing, 2005, pp 187–207 45. Lyons-Ruth K, Jacobvitz D: Attachment and disorganization, in Handbook of Attachment. Edited by Cassidy J, Shaver PR. New York, Guilford, 1999, pp 520–554 46. Swartz M, Blazer D, George L, et al: Estimating the prevalence of borderline personality disorders in the community. J Personal Disord 4:257–272, 1990 47. Stone MH, Stone DK, Hart SW: Natural history of borderline patients treated by intensive hospitalization. Psychiatr Clin North Am 10:185–206, 1987 48. Andrulonis PA, Vogul NG: Comparison of borderline personality subcategories to schizophrenic and affective disorders. Br J Psychiatry 144:358– 363, 1984 49. Akiskal H: Subaffective disorders: dysthymic, cyclothymic, and bipolar II disorders in the “borderline” realm. Psychiatr Clin North Am 4:25–46, 1981 50. Kroll J: Borderline Conditions. New York, WW Norton, 1988 51. Zanarini MD, Gunderson JG: Axis I phenomenology of borderline personality disorder. Compr Psychiatry 30:149–156, 1989 52. Stone MH: Long-term outcome in personality disorders. Br J Psychiatry 162:299–313, 1993 53. Keller MB, Shapiro RW: “Double depression”: superimposition of acute depressive episodes on chronic depressive disorders. Am J Psychiatry 139: 438–442, 1982 54. Maltsberger JR, Lovett CG: Suicide in borderline personality disorder, in Handbook of Borderline Disorders. Edited by Silver D, Rosenbluth M. Madison, CT, International Universities Press, 1992, pp 335–338 55. Russ MF: Self-injurious behavior in patients with borderline personality disorder. J Personal Disord 6:64–81, 1992 56. Leichsenring F: Quality of depressive experiences in borderline personality disorders: differences between patients with borderline personality disorder and patients with higher levels of personality organization. Bull Menninger Clin 68:9–22, 2004 57. Plakun EM: Prediction of outcome in borderline personality disorder. J Personal Disord 5:93–101, 1991 58. Stone MH: Borderline personality disorder, in Abnormalities of Personality: Within and Beyond the Realm of Treatment. New York, WW Norton, 1993, pp 215–257 59. Stone MH: The Fate of Borderline Patients. New York, Guilford, 1990 60. McGlashan TH: The Chestnut Lodge follow-up study, III: long-term outcome of borderline patients. Arch Gen Psychiatry 43:20–30, 1986 61. Greenberg HR: Fatal attraction: bring the lady a bug. Academy Forum 33:14–15, 1989


155

62. Linehan MM: Behavior therapy, dialectics, and the treatment of borderline personality disorder, in Handbook of Borderline Disorders. Edited by Silver D, Rosenbluth M. Madison, CT, International Universities Press, 1992, pp 415–434 63. Perry JC, Herman JL: Trauma and defense in the etiology of borderline personality disorder, in Borderline Personality Disorder. Edited by Paris J. Washington, DC, American Psychiatric Press, 1993, pp 123–140 64. Kernberg OF: The psychotherapeutic treatment of borderline patients, in Borderline Personality Disorder. Edited by Paris J. Washington, DC, American Psychiatric Press, 1993, pp 261–284 65. Snyder S: Pseudologia fantastica in the borderline patient. Am J Psychiatry 143:1287–1289, 1986 66. Lewis M: The development of anger and rage, in Rage, Power, and Aggression. Edited by Glick RA, Roose SP. New Haven, CT, Yale University Press, 1993, pp 148–168 67. Fingarette H: Heavy Drinking: The Myth of Alcoholism as a Disease. Berkeley, CA, University of California Press, 1988 68. Widom CS: Does violence beget violence? Psychol Bull 106:3–28, 1989 69. Rind B, Tromovitch P, Bauserman R: A meta-analytic examination of assumed properties of child sexual abuse using college samples. Psychol Bull 124:22–53, 1998 70. Simon RI: Bad Men Do What Good Men Dream. Washington, DC, American Psychiatric Press, 1996 71. Morse SJ: From Sikora to Hendricks: mental disorder and criminal responsibility, in The Evolution of Mental Health Law. Edited by Frost LE, Bonnie RJ. Washington, DC, American Psychological Association, 2001, pp 129–166 72. Halleck SL: The concept of responsibility in psychotherapy. Am J Psychother 36:292–303, 1982


CHAPTER

5 DEPENDENT PERSONALITY DISORDER AND HOMICIDE

A SEEMING PARADOX EXISTS regarding individuals who have dependent personality disorder (DPD): how can a person who seems so meek, gentle, and desirous of pleasing others commit such a horrendous act as killing another human being? When considered overtly, the two essential features of DPD individuals’ personalities—the pervasiveness of their dependency needs and their submissiveness—would not seem to predispose them toward violence, either against others or against themselves. What follows is a discussion of those aspects of the dependent personality that, under the right circumstances, can lead to a violent eruption.

FROM CHILDHOOD TO ADULTHOOD: ROOTS OF DEPENDENCY Although some argue that the personality traits consistent with DPD emerge only in young adulthood, the more pervasive psychiatric view is that the behaviors actually emerge much earlier and are noticeable in childhood by those who have contact with these individuals.1 There is also the possibility of some genetic factor operating in addition to social factors.2 Once a person with DPD reaches adulthood, the roots for the pattern of emotional attachments that he or she will develop have been laid. Powerful effects from the time when the person was more fragile and perhaps literally could not survive alone remain. The remnants of these cognitive processes persist into adulthood and lie ready to spring

157

158


to the ascendancy in complex relationships. Not only do these people feel they cannot survive without the person with whom they have chosen to have a relationship, but they also feel insignificant to others in their own right. Their behavior conveys that only through the person they have selected for such a special attachment role can they survive and have a meaningful sense of self. In adulthood, some individuals with DPD present a facade of independence. In fact, some achieve success in a particular vocation as long as their dependency is not disturbed. Others have handled their unresolved conflicts through the role of caring for others with similar dependency problems. In the jargon of the popularized mental health idiom, these people are considered to be “codependent,” meaning that their identity has become based on their taking care of other people or assuming responsibility for them. This behavior pattern conceals their own dependency conflicts. In many ways, however, this description oversimplifies the nature of dependency conflicts; it is based on an extension of the disease concept of addiction to more pervasive personality disturbances and character phenomena.3 However, the analogy to addictions is not without some merit, because dependent people often choose their mate or close friends on the basis that the other person will be available to them and “enable” them to remain in their addicted status.

PREDISPOSING FACTORS Let us now turn to some of the factors that can lead to acts of homicidal proportions by individuals with DPD. Their prominent needs to be cared for and nurtured are obviously not unique to them, and to a casual observer their lives often seem outwardly successful and uneventful. However, their difficulties arise when they feel their need to be loved is not being met and in how they react to that feeling. Many of the characteristics discussed in this chapter overlap and in reality cannot be separated; they are separated here only for the sake of discussion.

Inability to Free Self From Dependency A core conflict in DPD is the individuals’ belief that their well-being is necessarily contingent on others—often a particular person or institution. The absence of or betrayal by that person first engenders feelings of helplessness mixed with frustration. The frustration and anxiety occur in the context of a fragile security system that is easily upset. As a corollary, they experience an affective state in which they feel they are

Dependent Personality Disorder and Homicide

159

not in charge of their lives and lack an autonomous existence. With minimal coping devices to handle the specter of threatened losses, their selfesteem system remains under threat. When the devices to reinstate such a relationship do not succeed, depression or the use of addictive substances to palliate the anxiety and loneliness is often seen. As has been shown in social psychology and cognitive psychology, more benefits accrue to those who are able to exercise more choices in their lives and are not constrained by feeling compelled to carry out certain behaviors to maintain their security.4 Attempted breakups almost always meet with initial decisions to return to the relationship, although, at least on one level, the relationship is perceived as painful and destructive. Similarly, it is possible that the partner may be caught up in a reciprocal pattern on a different level. Then, power plays between the partners emerge in a desperate quest for a sense of power over the other in an effort to avoid the deep abyss of feeling alone and powerless. In the course of such relationships, the illusion is perpetrated that their relationship is deep, meaningful, and caring. On occasion, brief insights occur that not only do they not particularly care for each other, but they also desperately wish they could be free of the entanglements and destructive aspects of the relationship.

Case Example A 35-year-old white male had become attached to a woman from whom he felt he could not separate. Their relationship was volatile, destructive, and a source of pain for both of them, although he felt that his was the more painful role. He had met his partner when he was a patient in a chemical dependency treatment unit. The woman had also previously been treated for chemical dependency, but at the time of their meeting she was working as a counselor on the unit. They were immediately attracted to each other. Shortly thereafter, they arranged to see each other clandestinely, both on and off the unit, and soon became sexually involved. Following his discharge from the unit, their relationship continued. They both began to use drugs again, although she was able to keep her usage to a minimum level. Once removed from the institutional setting, the relationship began to show difficulties. The woman was demanding and controlling, and the man began to resent the demands she made on him. Rather than confronting her with her manipulative and extractive behavior, he conformed his behavior to her demands based on his fear that he might lose her. In time, her flirtatiousness and flamboyance in the presence of other men began to upset him. Although she seemed more mature than he and to have more worldly experience, she reminded him of girls he had known in his adolescence. Both individuals were of superior intelligence; he had pursued a doctoral degree for a time, and she had received superior marks in college before dropping out after 2 years because she became bored.

160


After 9 months, she told him she would be leaving in 2 days for a week’s vacation to think over their relationship. She did not tell him where she was going, and her relatives would not disclose her whereabouts. When she did not return in 3 weeks, he obtained the phone number of her residence in a different city from her former place of employment. When he called one night at 11:30 P.M., a man answered the phone; from the noise in the background, it was obvious that a party was being held at her home. He immediately hung up, but called again the next morning. She answered and informed him she did not wish to see him anymore and that he should not call. Contrary to her wishes, he began calling often, sent letters, and eventually announced that he was coming to see her. When he arrived, he found she had moved without leaving a forwarding address. He began to drink heavily, go on binges, and engage in high-risk activities such as driving at excessive speeds and becoming intoxicated to the point of unconsciousness. In desperation, he readmitted himself, heavy with nostalgia, to the unit where they had originally met. He remained preoccupied with her rejection, fantasized about her with other men, and dwelled on the humiliation of his experience with her. At that point, he decided he would find out where she was through her parents. He left the treatment unit and stopped in a liquor store to buy a fifth of liquor on the way to her parents’ home. He entered through a basement window, went to the kitchen, and waited for them to come home. When her father came into the kitchen, they shared a cup of coffee; the father then said he would take him back to the hospital. They talked about the past, which aroused the man’s nostalgia, with the twist that he recalled only the good times. When the mother came downstairs and saw him, she became agitated and insisted they leave for the hospital immediately. In response, the man announced that he would not leave until he had obtained their daughter’s phone number or address. They tried to convince him that she did not wish to see him and that he should return to the treatment unit. He grabbed a kitchen knife and held it up to their dog, with the ultimatum that the dog would die if they refused to give him their daughter’s address or phone number. When the father threw hot coffee in his face, he reacted by plunging a knife into the father’s chest and killing him. In attempting to intervene, the mother was also stabbed and killed.

In this case example, all the elements of pathological dependency were present, coupled with other signs of character pathology. Mixed elements, including sadomasochistic traits, were seen in both parties, although only one committed an act of homicidal violence. This case example illustrates how the victims of such violence need not be a primary person in the relationship but can be a third party or, in some cases, an innocent victim unbeknownst to either of the principals. In this particular case, the man, in his cringing inadequacy and inability to leave a relationship that was


161

doomed and painful, progressed along a path of personal disaster that eventually involved others. He felt incapable of living without the person who had become the recipient of his deep and unwavering attachments. He had never developed the ability to accept emotionally that one can survive various losses, build new relationships, and somehow, in time, survive for the better. Instead, he perceived an internalized, pervasive threat that survival was not possible without this other person, no matter what the price. This further illustrates the power of such attachments— much like a child who cannot thrive or live without the presence of the parenting object. Several years later he suicided in prison while serving his sentence for murder. Similar types of killings can also be ascribed to the “Othello syndrome.” In Othello, two ill-matched people with obvious incompatibilities feel a profound need for each other but cannot break off their relationship even though they sense danger. On one hand, a supposedly gentle and guarded young girl abandons her father to go with the Moor; on the other hand, the Moor, in late maturity, has always avoided such entanglements and believes he is risking his career by engaging in one now. Distrust and jealousy are fostered by his villainous military officer, Iago.5 Iago convinces the young girl, Desdemona, to establish her power by defending Cassio, Iago’s rival for Othello’s trust. Iago exploits Desdemona’s seductiveness, and while Desdemona senses danger, she cannot leave the situation and would not leave even if she could. When Othello falsely believes Desdemona is irretrievably lost to Cassio, he cannot bear her loss or tolerate his jealousy; he kills her—and finally commits suicide. Combinations of homicide-suicide cases often seem to meet this format in which the solution is to kill the person who is no longer willing and able to be in the facilitating role. The perpetrator has accompanying thinking that he or she must then suicide because he or she will then truly be alone—which leads to overwhelming anxiety. In one study of 20 such cases, 17 were intimate partners and all but one of the perpetrators was a male.6

Inability to Deal With Conflicts Some individuals with DPD use psychoactive substances to quell their fears of abandonment. However, for those who do not resort to substance use, being without the significant other creates a feeling of emptiness and of being at the mercy of one’s own feelings of fear, smallness, and insignificance. To conquer the feelings engendered by this state of helplessness and total inadequacy, some of the most brutal acts can occur in the form of a homicidal restitution.

162


Case Example A man had been involved for 12 years in what seemed to be a good marriage, based on public observations that he adored his wife. Yet behind the scenes, he held two jobs to meet their many financial and social demands. In time, loans also had to be taken out, which led to accumulated debt. Meanwhile, his wife engaged in the flamboyant lifestyle that she desired, with or without him. When she began to come home at 1:00 A.M. or 2:00 A.M., his need for denial of what this might mean was maintained by accepting her reasons that she had simply been having a good time with friends.

Up to this point, this example would simply have been one more marriage in trouble, and there would be no hints of the potential for a homicide. The interesting phenomenon in such cases is why, at a certain point, defenses break down. Sometimes, the final common pathway leading to the breakdown of defenses can be uncovered, but in other cases the reason remains a mystery. The case example continues: Eventually, the husband could no longer deny his suspicions about the veracity of his wife’s version of what she was doing when out. Reality testing began to intrude. A predisposing factor was the prolonged attempt to keep up their expensive lifestyle without relief. Playing into the situation was the wife’s expectation that things would continue in this vein indefinitely, a denial that the continued gaming she had engaged in would ever come to an end. When she failed to return home by the early hours one morning, her husband drove to bars and nightspots where they had once gone together or where she had gone alone in the past. At one of them, a bartender told him he had seen a woman who fit his wife’s description leave with another man. The husband waited at home for her return, at which time he asked where she had been. She gave the customary answer of being out with friends. For the first time, he confronted her with his belief that she had been out with another man. She responded, “So what if I was? At least I can have some fun with him and not be worried all the time like you are.” Within the space of a few seconds, he took a gun from a drawer and fired several shots at his wife, killing her. At the time of his subsequent psychiatric evaluation after the killing, he was filled with remorse and sobbingly told how much he missed his wife and what a wonderful person and wife she had been. He offered no explanation for what he had done and had trouble accepting that he was the person who had committed such a violent act. He did not recall firing the shots and could only remember taking the gun from the drawer. At the time of the homicide, this man was feeling abused and victimized; his repression and denial, usually operating, had temporarily broken down, allowing his hatred and aggression to emerge through his defenses. After the homicide, however, he returned to his customary behavior as a subservient and loyal person. The problem was that in this brief out-of-character interval, a homicide had been committed.


163

The weakness in the psychological defenses with which individuals with DPD operate over their lifetimes is seen in their inability to deal with conflicts that have often existed over many years. The postponement of dealing with the conflicts has allowed them to continue their seemingly well-adjusted lifestyles. Even apart from a homicide occurring, what they live through is a high price to pay for the inability to deal with the way their conflicts intrude on their lives. In this case example, the man’s image of servicing his wife to keep her happy, with minimal self-gratification apart from securing erratic dependency gratification, was overwhelmed when he could not reconcile his wife’s impinging disloyalty, dishonesty, and lack of appreciation for his dedication and suffering with the new evidence of her disloyalty. In many of these cases, after the homicidal act, the old pattern of defenses is reconstituted, coupled with the reidealization of the partner and their relationship. Except for the death of their partner, these individuals revert to the prehomicide status quo, with the difference that they often spend many years institutionalized. A serious dilemma posed by individuals with DPD is that they never know when their maladaptive defenses may fail; nor can this breakdown point be gauged much better by mental health professionals. These individuals’ self-esteem is so contingent on pleasing and waiting on others that they do not wish to face the anxiety and discomfort of confronting the difficult dilemmas that exist, which makes therapeutic efforts tenuous. When disputes between lovers arise, desperate attempts are made to continue the relationship at any price. In the last case, the price would have been for the man to accept that his wife was exploiting him, a price that would be intolerably high for the little affection that he received from her. Consequently, the issue of whether he was capable of leaving the relationship was raised. In the past, he had handled being victimized by the exploitative behavior of his wife by placing himself in a position of needing to deny his anger and resentment and to keep portraying himself as an individual constantly working against odds to make others happy, particularly the woman he loved. Although victimization served to distance himself from his mounting aggression and anger toward his wife, it simultaneously allowed him to maintain his self-image as a peace-loving person without murderous desires. His anger would be projected onto others to maintain this image.

Desire to Keep the Relationship at All Costs In cases where a joint murder-suicide occurs, the individual with DPD may have been motivated not only by the wish to destroy the person who disappointed him or her but also by a fantasy of the two (or more)

164


of them being together in some other type of existence, paradisiacal or not. Based on interviews with DPD individuals who survived suicide attempts subsequent to committing a murder, it does appear that their wish to be restored to perfect bliss with the person they had overidealized was one of their driving motives to the homicide-suicide attempt.

Deep Ambivalence A question arises as to why a person with DPD cannot simply act more rationally, give up the past relationship, and look for a new one. Given that the relationship has had a good deal of pain attached to it, this choice would seem like a logical step. Such a solution would also be parsimonious if available. Any person involved in an intense relationship is disappointed when his or her expectations are not fulfilled. Resentment and anger are inevitably experienced. However, it is the presence of deep and unresolved ambivalence that prevents a person with DPD from objectively examining his or her shortcomings in search for the answer to why a relationship with a friend, employer, lover, spouse, or other is not going well. Only when the ambivalence is not too prominent can the anger be expressed and the processes of detachment begin. However, when anger or hate surface in a relationship between a person with DPD and his or her partner, it is more difficult for that person to step back than it is for others. The lack of sufficient distancing keeps the intensity of his or her ambivalence at a high level.

Placing Blame for Unhappiness on Others Individuals with DPD find others to be at fault if those others do not respond in the manner the individuals demand; DPD individuals hold the inculcated view that the others (“need satisfiers”) exist for the purpose of meeting their needs when the occasion arises. The potential for more violent behavior becomes higher when DPD individuals shift their thinking to believe that others not only are at fault but also must pay a price for disappointing them, even to the degree of needing punishment. In some cases, their internal conflict progresses to a debate about whether suicide, homicide, or a homicide-suicide finale is the needed course of action. When an individual with DPD attacks others—physically or verbally—the implication is that others are responsible. Impulsive acts of rage and anger to manipulate others are especially possible for these individuals when comorbid states of histrionic or borderline personality disorders are present. Wanting desperately for others to take over and alleviate their pain makes their frustration with accompanying anger inevitable. The quest to blame others also reflects how aggression looms


165

under the surface of dependency. To cling to another in a demanding and unremitting fashion is an assertive act. Seeming dependency does not mean the person cannot mobilize his or her aggression to maintain a relationship.7 Demands for nurturance in his or her everyday life can be read as aggressive acts. The persistent illusion is that others can solve the DPD individual’s problems and, by their not doing so, are responsible for his or her misery. The typical situations that raise these conflicts to homicidal levels are seen in marriages that are coming apart, love relationships threatening to end, betrayal in business or personal relationships, prolonged divorce negotiations where attorneys or judges may be blamed as well as former partners, custody disputes, or divorce settlements in which one party seems to get an unfavorable distribution. A psychiatrist or some type of counselor in a treatment or counseling relationship might also be on the receiving end of blame for not restoring a relationship to the “old happy days.” How do the frustration and resentment finally rise to a homicidal level in these settings? Conversely, why do not more people commit homicides in such emotional dilemmas? A beginning answer is that individuals with DPD attack others in ways that are to some degree the equivalent of homicide far more frequently than is appreciated—sometimes the attacks are physical, but more typically, they are verbal or through social manipulations that incur the wrath and indignation of the others. Lawsuits are one device for attacking someone else and securing the support of others—that is, those people whom DPD individuals pay for help, such as attorneys or experts. In turn, such attacks may lead the party being attacked to respond by counterattacks or countersuits. In some of these situations, the anger is displaced onto mental health workers, attorneys, or judges. For example, in one homicide case, through 3 years of futile divorce negotiations, the husband consistently blamed his wife’s attorney for the couple’s marital difficulties. The husband eventually shot the attorney, thereby continuing to deny that his wife had been the one who first moved for a divorce and who had hired the attorney. Consider the reflections of a man who had killed his wife and daughter sometime after his wife told him she was leaving him. He had come to realize that his pleading efforts to avoid her leaving him had failed. Two years after the homicides, he reflected: I can’t understand why I did it. In fact, I still sometimes doubt it was me who did it. I’m not someone who kills people. It’s like she’s still with me. Why did she have to try and leave? She treated me badly. I was embarrassed by her a lot…yet we had such wonderful times together. I keep thinking over the good things we did and of what our future would have been like. How can I stop thinking of her? I still dream about her and wake up thinking of her.

166


This pattern of individuals with DPD attacking others, based on their own unhappiness and frustration, is not uncommon. The attacks reveal their ambivalence because, in one way, they reflect distorted relationships that have become altered if not lost. It is important to note that more than serving as a means of revenge, the attacks on others through formalized mechanisms, such as lawsuits, dragged on for years on appeals, are another way of prolonging relationships. It is a matter of prolonging the relationship one way or another, and if it needs to be done via legal entanglements, it keeps the respective parties involved and avoids a total separation. In that manner, the self-recriminations can also more easily be directed against others. After such situations have ended, a suicide or homicide becomes more likely. Attacks on a more conspicuous level are these individuals’ attempts to alleviate their own distress in a more easily understood manner.

Case Example A 32-year-old male who had only had sporadic employment for a few months at a time was living separately but in a house owned by his parents. Although a college graduate, he had never taken steps beyond applying for graduate or professional schools. He had been accepted in several programs but either turned them down or got yearly postponements until dropped. The parents had continued to provide for his style of living, allowing him to live in a middle-class neighborhood. His spending indulged in some status items but was not otherwise ostentatious. A few scattered clinical contacts, whom he saw at the insistence of his parents, gave a personality disorder diagnosis. Over the years he kept insisting, in response to his parents’ “intruding,” that he did not want his parents interfering in his life. When they pointed out that his style of living was totally dependent on them, he would become angry and distance himself for months. During these intervals, the parents’ support continued. When the father died, the financial situation changed, and his mother was advised to sell the house the son was living in. In an angry confrontation during which he accused his mother of betraying him, he assaulted her and beat her into unconsciousness, which led to a charge of attempted murder.

Need for Immediate Response The relationship problems experienced by individuals with DPD are multiplied by their inability to tolerate periods of waiting or ambiguity. Their demands for an immediate solution to their relationship problems or to be restored to a sense of homeostasis by the presence or efforts of others alternate with the threat that they will become immobilized by the onset of depression, if this regression has not already occurred.


167

Depression and Denial As noted, individuals with DPD may experience occasional depressive episodes when their feelings of helplessness are prominent; however, for most of their lives, they are not seen as clinically depressed by others or by themselves. Their reactions are rather of the situational variety with depressed affect. Because these individuals assume subservient roles in some part of their personal lives, they often function at a much lower level than they are capable of, in terms of both their personality attributes and intellectual capacities. Hence they may be underachievers in some major areas, even though they are talented. To maintain a dependent relationship may mean thwarted ambitions by abandoning plans or promotions that require leaving a dependent setting. They often lack a sense that they are important and worthwhile in their own right or that they can exert a significant influence on how others behave toward them. Many live lives of quiet desperation, simply hoping not to have their structured dependency disrupted. They often feel frightened, if not terrified, at the possibility of losing their support system within their family network. This corresponds to their belief that they have little influence over having their basic needs met in relationships with others. Blatant denial may serve to keep them functioning until a massive disruption looms.

Anxiety and Obsessional Symptoms When the dependency techniques are not working, individuals with DPD often display related behaviors that reflect anxiety disturbances and obsessionality. One sequence of obsessional behavior may involve difficulty in making a decision and seeking advice from others about it—but then, after making the decision, retracting it and seeking reassurance from diverse sources that it was the right one. Other anxietyridden characteristics include being uncomfortable when alone and trying diverse ways to avoid being alone or engaging in maneuvers to extract support from others. These behaviors surely reflect an individual in a state of conflict, but they do not serve as predictors of a homicide.

Comorbid Disorders The high incidence of a dual diagnosis can complicate efforts to understand DPD. Comorbid disorders may include substance dependence or abuse, major depressions, bipolar disorders, or diverse anxiety disorders. Also, overlap with other personality disorders (e.g., histrionic,

168


schizotypal, narcissistic, avoidant personality) is often found. Given this diversity of comorbid diagnoses, it is a rare phenomenon to find a person with DPD without at least one more diagnosis.8 Another variation is that the person with DPD may develop actual physical pain or disability with somatization processes. Chronic fatigue syndrome, a diagnosis with many vagaries, may fit many of these people.9 In one case, after months of staying in bed for almost 24 hours a day, an individual emerged with his old-style zest and energy. This unfortunately led him into confrontations and eventually into what he believed was a justified homicide against someone who had earlier wronged him. The significance of drug and alcohol abuse can also be gauged in the context of such dependent personality problems. The chemicals can help maintain, for a period of time, an illusion of being attached and not alone.

BEHAVIOR PATTERNS WITHIN DEPENDENT PERSONALITIES THAT CAN LEAD TO VIOLENCE In examining the DPD individual’s more covert behavior patterns, just how some of these behaviors can lead to a violent act become clearer. These individuals’ dependence on others, coupled with their fear of being alone, sets the stage for them to experience major anxiety episodes coupled with fears of abandonment. On one level, they engage in defensive maneuvers to ward off the possibility of being alone. Ingratiating behavior with others or attempts to “bribe” others into continuing relationships are typical ploys characteristic of their lifetime patterns of tenuous adjustment. Their fears are not confined solely to losing love when a relationship is severed; instead, their work efficiency and their entire capacity to continue functioning are compromised when a dependent relationship is threatened or not working smoothly. Individuals with DPD go to great lengths to prevent a rupture of relationships, be they love relationships in a romantic sense or relationships meeting some other need (e.g., general affection, acceptance, attention, praise). Unfortunately, their sensitivity to an anticipated rupture of relationships is often so high, and their fear of loss so great, that they begin to take defensive and anticipatory measures early, which then become expansive and exaggerated. In general, many of these people form attachments in what they believe at the time are love relationships. At first, the relationship seems


169

to promise meaning and new significance in their lives. They believe these new relationships will be the “cure” for their loneliness and their sense of being lost or of not connecting with meaningful relationships. In such a situation, all the hallmark characteristics, if not the addictive qualities, of a dependent state are seen. In a broad sense, these distorted attachments and love relationships may seem so commonplace that observers can easily miss some ominous potentials of what may emerge if the dependency relationship is altered. Once DPD individuals are involved in this kind of relationship, it is almost impossible for persons with DPD to extricate themselves because they fear a return of their deep loneliness. This loneliness is not perceived as being transient but rather as something overwhelming to them, even to the point where they feel they will not be able to survive if the relationship ends. The thought of this significant relationship ending induces a high level of anxiety, or even panic states, in which they are devastated, helpless, and hopeless. Stalking may occur in some cases when a fantasized dependency relationship has developed in response to loneliness.10 The fantasy takes the form of an attachment to a person who may never be aware that he or she is the object of the fantasy. Some individuals progress to an erotomanic delusional state. Overt steps to maintain the relationship occur through stalking, which again may go on without detection for some period of time. However, when the stalker becomes more aggressive in his or her need for contact by way of letters, e-mail, phone calls, sending gifts, or making appearances, a threshold has been passed. These last steps often arise because of heightened dependency needs or fears that the stalking at a distance cannot give the DPD individual the reassurance he or she needs. Persistence and aggressiveness may increase. Efforts to end the stalking push the individual to the point of viewing this as ending the relationship. Because the psychiatric status of the individual is unknown to those intervening, they may not be anticipating a homicidal attack, which can occur against the primary object being stalked or those seen as trying to end the stalking. In the DPD individuals’ precarious state of trying to assure themselves that their needs will not be left perpetually unmet, a regressive shift may occur wherein individuals with DPD begin to sense that their dependency situation cannot continue as it has been. The realization can emerge slowly or precipitously. The experience appears to be one in which they sense either that the emotional protection of a relationship could be lost or that the approbation received from a key relationship will no longer be forthcoming. Furthermore, their submission or dedicated service, if not obeisance, no longer works. Even credible achieve-

170


ments in an objective sense do not suffice to reassure them, because this reassurance emanates overwhelmingly from others’ praise. At worst, the slavish dependence on others for a sense of self-esteem has built-in limits of having to live in an atmosphere of avoiding conflicts or competitiveness. Pushed to an extreme, persons with DPD may display masochistic personality traits; the DPD individual’s avoiding confrontation and seeking ways to submit to others allows these others to gain the ascendancy. Individuals with DPD do not feel capable of carrying out separations that might otherwise occur at this stage. Their symbiosis negates that degree of freedom. They need to remain attached despite how punishing and painful the relationship has become and regardless of the strivings they feel to become more free and autonomous. At that point, some may also resort either periodically or regularly to using drugs or alcohol. When a relationship shows signs of breaking up, individuals with DPD first try diverse ways to prolong the relationship. They may engage in desperate measures or pleading to entice the other person to stay involved, or they might withdraw from the situation in an attempt to gain some distance from the other person involved in the relationship. By so doing, they hope that they can expand their repertoire to include new ways of gaining reassurance. However, the odds are that in time, they will fall back on their more traditional patterns of ingratiating and placating others or, as a variation, doing tasks or “dirty work” others will not do, with the hope and expectation that a sense of tranquility and jaded confidence will return from their extra efforts. When they then shift back into their pleading or possibly threatening maneuvers, it is an indication that their renewed efforts are not working. They are now approaching a danger point for possible violence.

Defects in Ego Functioning Often individuals with DPD are seen as being inadequate people because of their conveyance of helplessness; however, that explanation is too simplistic for understanding how something as complex as a homicide can occur in their personal lives. Individuals with many of the other personality disorders also have types of inadequacies that may contribute to a final common path toward a homicidal act. However, those in the DPD group have certain characteristics in their sense of self that accompany the defect. Their problem with self-esteem has been operational throughout their development and into adulthood and becomes the antecedent vulnerability that leads them to an ultimate homicidal act. A certain combination of circumstances elicits the final state of helplessness.


171

Many defects in basic ego functions are present, which is what leaves these individuals vulnerable to spiraling into a homicidal state and thwarts them from acting on the basis of their sensed needs. Their dependence, with the inability to step out of such destructive relationships, reveals a limited perspective and narrow focus to their existence. Although some refer to this vulnerability as an impairment in reality testing, it does not reach the level of psychotic thinking. Rather, it is a prominent awareness of the limitations existing in the degrees of freedom that these individuals have in their lives. For those who also have some borderline personality features, impulsive acts and exercise of poor judgment start to appear; they begin to portray themselves as not having control over themselves or they misinterpret therapeutic help. They have a sense of being caught in the grips of powerful forces with which they cannot deal. Even more ominous is a situation in which a person with DPD is not aware of the degree to which he or she is distorting the assessment of the person with whom he or she is involved. He or she feels unable to delay, reflect, and wait for resolution of painful situations because he or she does not wish to endure the pain, anxiety, and loneliness of these options. One variation that occurs with DPD individuals is when the reality of a loss arises because the partner is terminally ill. Such a situation may lead to a murder-suicide combination. In some cases, a mercy killing occurs but the body is kept preserved, such as in a freezer, to maintain a connection. In time, a correlative amount of aggression will accumulate within the DPD person and induce a state of discomfort. Yet the person is unable to express his or her excess anger openly because by so doing, he or she would risk being further pushed into the precarious state of being abandoned. The situation of having to choose between expressing anger, with the accompanying threat of possible further loss of support, versus doing nothing and feeling helpless to continue living one’s daily life cannot persist indefinitely. Earlier clinging attempts have pushed a relationship further apart. As noted, a person so conflicted may shift into a depressive episode with a full emergence of a major depressive disorder and increased suicidal possibilities. Another possibility is that the DPD person’s anger and rage will lead to more impulsive behaviors. The latter may occur in attempts to coerce rescue efforts or to punish others, especially the one he or she sees as being the source of pain. The DPD person may begin to dwell on how to hurt the other person on an equivalent level to what he or she feels. A form of pseudocontrol results over his or her unresolved aggression. Rather than turning in new directions, which would imply a

172


resiliency in seeking new sources of gratification, the dependent individual is fixated on one or a few limited devices. Hostile feelings, ambivalence, and thoughts of revenge predominate, with regressive efforts to gain security, as in the following case example.

Case Example A middle-aged man’s wife of 25 years left him a note saying she was leaving him and moving to another part of the country. After her departure, the man began to spend evenings with his widowed mother in a residence for the elderly. While there, he would read the newspaper, much like he had every night with his wife. After several months, his behavior switched into another level, with him dwelling on why she should be entitled to go off and lead a happy life while he suffered. Eventually he tracked her down, stalked her, and killed her in the midst of an argument.

HOMICIDE IN THE CONTEXT OF MARITAL RELATIONSHIPS One of the first people to study in depth the psychodynamics of homicide committed in a marital relationship was Bruno Cormier.11 His investigations were not confined to those homicides that were labeled as a product of a mental illness or motivated by material gain. Cormier sought to get away from focusing on the individual’s reactions either at the time of apprehension for a murder or at the time of trial, when the more immediate or early consequences of a homicidal act become prominent. At trial, a contemporary emotional state (e.g., jealousy, infidelity, adultery, alcoholism, marital violence) often becomes the focus of inquiry as one of the usual precipitants of a killing. What Cormier wanted to do was explore how a homicide could occur in the ordinary context of ongoing love relationships in which friends and acquaintances of both parties had not seen them as being much different than anyone else. To pursue this project, he followed a group of men for 2–8 years after they had been convicted of killing either their wife or a woman with whom they were involved in a committed relationship. The key to each of these men was a pervasive pattern of unresolved dependency conflicts. Similar formulations could be offered about women caught in these dependency predicaments. In studying his cohort of men, Cormier observed that irrespective of their social class, a strong tie existed between the partners in all of these relationships. Each partner realized that he or she needed the other for help in solving their mutual problems. Mixed with the good times were disruptions based on occasional distrust, jealousy, or need to retaliate for


173

a hurt. Yet despite a heightening of the difficulties to the point of overt aggression against the partner, even to the point of threats to kill, no longterm separation between the partners occurred. However, at some point, one of the partners’ attachment for the other began to weaken. Perhaps the emotional cost of the relationship had finally taken its toll. Parting at that point would seem logical and, with time, an imperative, but it often does not occur. The inability to start the processes of loosening the dependency tie and facing reality is a crucial variable if a possible catastrophic ending is to be avoided. At a certain point in the natural history of these “normal” homicides, one party begins to realize that although he or she cannot seem to separate from the partner, he or she cannot go on indefinitely in the pattern that exists either. Both partners are trapped with each other as if by fate. In some cases, it is revealed that one or both partners began to have preoccupations that killing might be the solution. Eventually, at a point of intense emotion, when continuing the relationship is inconceivable but separating also seems impossible, a homicide occurs.

PERPETRATORS’ REACTIONS FOLLOWING HOMICIDE Perpetrators’ reactions to having killed someone can vary in cases in which they have unresolved separations. Although some may then attempt and succeed in committing suicide, those who survive such an attempt continue to be preoccupied with death for many years. Some have a sense of loss, like an amputation of a part of their selves. At times, this sense of loss can produce transient periods of relief, in that the person is denying the finality of what has occurred. Yet denial that anything has happened cannot persist very long. Many of these individuals follow a pattern of first blaming the lover almost exclusively for all the problems before the act of killing occurs but after the killing seeing themselves as the only guilty one and entirely at fault. In the great majority of these cases, their guilt is upheld in the legal system, even though they may not be convicted of first-degree murder. Another change in their perspective after the homicide is that rather than seeing that there was no solution to their problems, as they did before the homicide, they are now able to appraise the diverse possibilities that were present and why these were not considered. When an individual with DPD reaches that junction, a course of genuine remorse and sorrow can begin. The completion of that course can be accomplished only when he or she has accepted the reality of the loss, after which the processes of painful detachment can finally occur.

174


Another possibility following a homicide committed by those with DPD may be that they experience a state of depersonalization and derealization as a means of dealing with their attendant anxiety. Some describe themselves as having observed the act or being numb to any feeling at the time of carrying out a homicide. The crux of a homicidal situation is that the person with DPD has put up defenses to the degree that he or she is not capable of experiencing his or her performance of the act but rather views it as if witnessing someone else doing the deed. The DSM-IV-TR1 diagnoses of depersonalization disorder and dissociative trance disorder will undoubtedly become quite popular in the legal arena with such cases.

THERAPEUTIC CONCERNS When caught in a relationship from which they are powerless to extricate themselves, individuals with DPD may seek counsel from others about what to do. Those giving advice may direct them to be more selfassertive, carry out other commonsense acts to make them feel better, or simply express their aggression. Unfortunately, expression of their aggression does not assuage the conflicts related to their dependencies. It may actually make matters worse by heightening a sense of failure and guilt when, having followed the advice of others, they do not begin behaving in an “improved” manner. For example, one man had been a member of therapy group where materials on learned helplessness were read. The hope was that by understanding helpless states, the group members could begin to resolve them. This individual, however, when released from the hospital on a pass, killed his lover. He later expressed thoughts that he was a poor learner. It is instructive to reflect on cases of misdirected therapeutic efforts where the dependency needs of these people are capitalized on and even accentuated by therapists in the hope that the catharsis will be therapeutic. Some therapists who have pursued this approach have at times promoted vengeful acts, such as the filing of lawsuits, in the belief that these acts will reflect autonomy and promote independence. Therapies that encourage the direct expression of aggression are based on the idea that patients will then not feel as helpless or used by others. What is missed in these approaches is whether the person has enough ego strength to contain and direct his or her aggression without violence. In these miscarried therapies, the repair of a damaged personality has only been postponed, and it may take much suffering before the person realizes that nothing has been solved within his or her own personality by simply mobilizing anger in this manner.


175

CONCLUSION Inquiry into the background and developmental factors leading to a dependent-type personality is always relevant. Some factors allow a partial understanding but still leave mental health professionals largely unable to prevent the violent outcome that occurs—fortunately, only in a few of these silent, suffering personalities. In essence, dependent persons remain in situations based on their self-image of serving others and with guilt over too much assertiveness or aggression. Ordinarily their guilt is tempered through the punishment they receive from others. In the final outcome, the homicide appears to be carried out via an ego-splitting mechanism. One part of the person carries out certain acts that are partially dissociated from totally conscious control and rational functioning. In some of these cases, defenses may be loosened by alcohol or drugs. In others, the background of stress over years of unresolved conflict prevails. These individuals routinely use a great deal of denial and repression, but the weakness of these defenses is that they permit these individuals to remain distanced from their deeper feelings of anger. At a certain point, the potential for a homicide is reached. Dissociation that allows a homicidal act is one possibility. The direct breakthrough of all the hate and anger, from which they later attempt to dissociate, is what occurs— irrespective of some amnesia that may occur later for parts of the violence committed because of its inconsistency with their self-image.

REFERENCES 1. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 2. O’Neill FA, Kendler KS: Longitudinal study of interpersonal dependency in female twins. Br J Psychiatry 172:154–158, 1998 3. Hauken J: A critical analysis of the co-dependency construct. Psychiatry 53:396–406, 1990 4. Bandura A: Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ, Prentice Hall, 1986 5. Wangh M: The tragedy of Iago. Psychoanal Q 19:202–212, 1950 6. Felthous AR, Hempel AG, Heredia A, et al: Combined homicide-suicide in Galveston County. J Forensic Sci 46:586–592, 2001 7. Bornstein RF: The Dependent Personality. New York, Guilford, 1993 8. Loranger AW: Dependent personality disorder: age, sex, and Axis I comorbidity. J Nerv Ment Dis 184:17–21, 1996

176


9. Bock GR, Whelan J (eds): Chronic Fatigue Syndrome: Ciba Foundation Symposium 173. New York, Wiley, 1993 10. Mullen PE, Pathe M, Purcell R: Stalkers and Their Victims. New York, Cambridge University Press, 2000 11. Cormier BM: Psychodynamics of homicide committed in a marital relationship. Corrective Psychiatry and Journal of Social Therapy 8:114–118, 1982

CHAPTER

6 NARCISSISM AND HOMICIDE

TO UNDERSTAND, beyond a purely descriptive level, how individuals who have narcissistic personality disorder (NPD) can become engaged in violent behavior, some type of theoretical perspective to comprehend narcissistic personality functioning is needed. When the diagnostic criteria for NPD are considered, the potential for various factors to go awry in these individuals becomes clearer. More generally, there is the question of what role a narcissistic core may play in several personality disorders and whether narcissistic injuries are one key to unraveling why some of these individuals become homicidal. What is always intriguing is how so many individuals with narcissistic personalities are quite successful in their lives, be it in artistic endeavors, academia, or business pursuits, whereas other narcissistic individuals spiral into endless problems in their interpersonal relationships. Some of those in the latter group are candidates for committing a serious act of violence. It is only in retrospect that the reasons for such violent acts can be suggested.

CHARACTERISTICS OF NARCISSISTIC PERSONALITY DISORDER Interrelated Feelings of Grandiosity and Unworthiness Under the DSM-IV-TR1 criteria for NPD, one hallmark of the disorder is a pervasive pattern of grandiosity in fantasy or behavior. Individuals with NPD feel the need to exaggerate their accomplishments and talents and expect to be singled out as special—even in the absence of corresponding achievements. Even for those whose talents merit applause,

177

178


whatever is forthcoming rarely is sufficient for their needs. Correlated with this grandiosity is their difficulty in assessing their actual talents or contributions over the long run. The stage is then set for progressive disagreements and conflicts with others. Despite their own sensitivity they lack empathy and cannot recognize or identify with the feelings or needs of others; hence their self-absorption and self-centeredness. NPD also has a close connection to antisocial personality disorder, which can provide one pathway for potential antisocial or violent behavior. In one study, those with NPD scored even higher on grandiosity than those with antisocial personality disorder.2 Work on juvenile psychopathy may unfold insights into juvenile narcissistic development as well. Another situation that may occur is when a person with NPD is unable to come to terms with a situation in which someone else may be more unique or accomplished in a competitive area than he or she. These individuals begin to view the person who is thwarting their own advancement as the chief interferer with the fulfillment of their narcissistic wishes; in this sense, this outside object (the other person) is seen as a flaw in a reality that is already distorted. Closely allied to the feeling of self-importance is a feeling of unworthiness. In a superficial sense, this may seem paradoxical. However, an examination of these individuals’ feelings of worthlessness and inferiority eventually uncovers a core of diminished self-esteem. The result is that even the genuine achievements that accompany their talents often leave them plagued with a sense that they are fraudulent. As a consequence, any legitimate pleasure that should result from their accomplishments is compromised. Instead, they remain preoccupied with how well they are doing and their need to be thought of highly by others. NPD individuals’ need for attention, affection, and admiration seems boundless, and when these are not forthcoming on a regular basis, they resort to manipulative behaviors in an effort to extract the necessary positive feedback, no matter what the price of such behaviors might be. The lack of feedback can range from a lack of compliments to the many disappointments that routinely occur in life. Slights, such as a tone of criticism by another person or not being placed in a position of special importance in some interpersonal relationship or at work, can become sources of brooding. Rage, shame, guilt, and humiliation emerge subsequent to these perceived slights, although they may be hidden under an attempted exterior of indifference. In the course of seeking relief for this pain, individuals with NPD can become exploitative by using others to aggrandize themselves. They may become involved in criminal activity—even violent behavior—to accomplish their goals of reestablishing a sense of regained power. For example,

Narcissism and Homicide

179

they may engage in fraud and embezzlement to enhance their self-esteem and restore their sense of undiminished power. Alternatively, they may initiate schemes to destroy some other person, often one whom they see as standing in their way of gaining even greater glory. These dynamics have been witnessed in the collapse of various prestigious investment banks, in which individuals’ quest for power, played out over large stakes, merged into criminal activities despite already glamorous and successful lives.3 Similarly, the exposure of deceptive, if not illegal, practices of high-level officers in corporations that has led to criminal charges in some cases can be viewed as related to a sense of unlimited narcissistic entitlement. On one hand, their sense of entitlement blinds them from seeing the unreasonableness of their demands and behavior. On the other hand, their grandiosity contributes to what they feel justified in doing. Hence, they have a sense of immunity for their actions. The same dynamics operate with more minor and petty offenses, reflected in personal violence as a way of bolstering sagging self-esteem. Despite what they achieve through these offensive behaviors—whether they be legal or illegal—individuals with NPD still have a constant and pervasive sense of feeling deprived and burdened by seemingly insatiable inner needs, which continues to attack their self-esteem. One unstable solution to fulfilling those needs is to merge into idealized or strong authority figures and seek narcissistic sustenance by such a connection. However, merging neither contributes to a person’s sense of cohesion of the self nor lessens a person’s sense of fragmentation and depletion. To the extent that self-cohesion suffers from these attempts to merge, efforts are directed at ways to regain a sense of narcissistic tranquility. No matter how these individuals may fail in attempts to bolster their self-esteem, their expectation persists that they are special and therefore entitled to something more than others—and more than they themselves are receiving. If illegal acts are resorted to, it is not seen as a situation requiring punishment but rather as an entitlement. Their selfesteem is then maintained by their being an “exception” to the others, one to whom others owe the fulfillment of their expectations and fantasies. If such expectations falter or do not come through, progressively desperate attempts may be made to change the situation. This is the background that can give rise to more desperate antisocial trends in some, even to the point of homicidal violence.

Lack of Insight Given their exquisite alertness to how others treat them, there might be an expectation that individuals with NPD would be sensitive to the

180


emotional needs of others. Surprisingly, their characteristic self-centeredness makes it difficult for them to gauge how others are reacting to them and to have the empathy necessary to recognize the emotional needs or states of others. NPD individuals’ empathic impairment is related, in turn, to their sense of entitlement, manifested in their periodic grandiosity that makes them impervious to realizing, or wanting to realize, how others may react to them or their blameworthy behavior. The result is a lack of insight about many of their own feelings.

Blaming Others The phenomenon of blaming others is an interesting question that can only be answered based on an understanding of diverse types of human developmental processes. At a certain stage in their development, children begin to make attributions that others are at fault for their pain or discomfort. Blaming is seen as occurring early in infancy by way of projecting inner pain to some outside source. It has been theorized that blaming begins much earlier than the more developed model of not blaming others and taking responsibility for one’s own discomfort. The latter requires a more developed sense of reality testing along with selfappraisal.4 Blaming others does not get beyond the level of disavowing and attributing to others part of oneself. For individuals with NPD, blaming becomes a pervasive part of their personality functioning beyond the function it serves in others. Blaming becomes allied with NPD individuals’ other narcissistic components in their need to attribute fault to others. Again, this additional strand contributes to a lack of responsibility for what the person does.

Experiencing Depressive Symptoms The predisposing characteristics of grandiosity and low self-esteem, with their accompanying vulnerabilities, create problems in individuals with NPD similar to the diatheses in those with borderline personality disorder. Both types of people are known to be particularly vulnerable to depressive episodes. The narcissistic-based type of depression seems more acutely related to fluctuations in self-esteem, in contrast to the fragmentation and regressive types of depression witnessed in borderline individuals. In a sizable number, if not the majority, of individuals with NPD, the depression is often not experienced in a full clinical sense but rather as a brief situational reaction. Such reactions may be attributable to the functioning of those personality traits striving to ward off a major depression. If these individuals begin to believe that fate has turned against them, but manage to avoid entering a major depression, they may slip


181

into the role of being a victim. However, the victim role does not persist as long, nor is it as pervasive, as in those who have borderline personality disorder. Resorting to manipulation and exploitation devices allows the narcissistic individual to maneuver around the sense of weakness and powerlessness that goes with extended victimization. The unresolved fragility in self-esteem is what makes those with NPD perpetually vulnerable to some degree of depression. Because of their unstable relationships with those to whom they are close, an intermittent state of depressive affect never quite goes away. Some narcissistic individuals are quite adept at using the victim role to set themselves apart from those who do not portray themselves as victims. By so doing, they gain sympathy from others and allow themselves to feel temporarily superior. Also, by portraying themselves as the ones who have been deprived and misused, they reverse roles with those who may have victimized them in the past. In fact, at times narcissistic persons may switch from being the exploiter to being the one who is exploited when the former role fails to gain them success in a narcissistic battle. However, if they stay in the role of a victim too long, or become unable to manipulate their way out of continuing disappointments, they may experience a major depressive episode. This sense of despair emerges when they have failed in their power game to manipulate and dominate and have depleted all their manipulative resources. In this more desperate state, the potential for some violent solution arises.

Paranoid Behavior It is not appreciated how often pathological degrees of narcissism may leave an individual vulnerable to paranoid disturbances. Such an ominous turn can be related to a sense of misplaced trust, in which these individuals feel they have been betrayed, or when brief insights emerge about their own integrity and questions arise about what they stand for.5 In the latter case, their feeling of being adrift induces an uneasiness within them, accompanied by increasing confusion and hypervigilance. Some take the final regressive step into a persecutory system with delusional ideas, which allows their grandiosity to survive, but at a great sacrifice.

NARCISSISTIC RAGE A narcissistic type of homicidal act arises within the matrix of an archaic type of narcissism, referred to as the phenomenon of narcissistic rage. As noted, narcissistic individuals may develop an indifferent or cool exte-

182


rior as an initial response to threats to their self-esteem. However, when their composure gives way, it is striking to see the intensity of their anger and need for revenge. Theoretically, narcissistic rage involves themes of grandiosity and power; psychodynamically, it involves archaic and omnipotent object representations that have become internalized within the person’s psychic functioning. Within this explanatory framework, narcissistic individuals commit acts of homicidal aggression with an absolute conviction that the acts are needed. The underlying motivation is to prevent themselves from regressing into infantile states of helplessness, if not psychosis. The threat is to the illusion they maintain of being totally self-sufficient and in control of people or situations—an illusion that contributes to their lack of empathy. In a study of offenders with a history of very serious offending, significant associations were found between NPD and homicide.6 When in prison, narcissistic individuals were intolerant of rules and regulations and believed that violence was the only solution for their interpersonal difficulties. The reality of prison discipline led to diverse problems because it clashed with their ideas of how they should be treated. Narcissistic rage encompasses a broad spectrum of clinical experiences. It is not confined simply to an act of homicidal proportions but is best visualized on a continuum. Many trivial occurrences of rage occur in states of annoyance or irritation or when someone is aloof or does not reciprocate expressions of warmth to cover the potential for the narcissistic person’s rage. At the other end of the continuum, narcissistic rage encompasses serious outbursts, such as the furor expressed in a violent attack carried out far beyond what is necessary to achieve a killing (e.g., continuing to beat an already dead body). Again, this rage exemplifies how one killing is actually being carried out multiple times. In diagnostic terms, narcissistic rage reactions can also be seen in catatonic-type reactions or paranoid delusional systems as well as in depressive episodes connected with a depletion of self-esteem. Hence these rage reactions are not confined to certain types of personality disorders. Narcissistic rage encompasses areas of aggression, anger, and destructiveness in general. Using the concept of narcissistic rage based on a sense of self-righteousness, Horowitz7 illustrated how the narcissistic individual’s reaction to these perceived assaults can lead to violent actions. Selfrighteous rage is a state of mind that arises most frequently during narcissistically vulnerable stages of development. As such, it is more prevalent in those who have experienced pathological twists in their narcissistic development. It is also seen as an emotion that develops in


183

the context of other affective components, such as becoming integrally entwined in a cycle of rage and anxiety. The narcissistic individual, underneath a facade of control, is often left with a chronic sense of embitterment. A withdrawal into a type of apathetic dullness may slowly take place. If the narcissistic individual is seen at that point in clinical settings, a descriptive diagnosis of some type of major affective disorder is often given and some type of antidepressant medication instituted to alleviate some of these symptoms. In a longitudinal and historical sense, this type of depression is coming at the end stage of these developmental antecedents. When some narcissistic individuals reach this state of threatened depression, they may attempt to flee into the opposite emotional state. Analogous to manic behavior, the opposite state does not portend less risky behavior. For example, instead of exploding in self-righteous rage, they may pump themselves into a state of buoyancy and seem on top of things. Their sense of entitlement may increase. Instead of experiencing shame, rage, and anxiety, they may experience a sense of excitement or creativity in different areas of their lives. Rather than acting out chronic embitterment, they may shift their behavior toward heightened social participation in an effort to obtain attention or praise. Women incarcerated for violent crimes also have been seen as having an association with NPD.8 Underlying these threatened states is the potential for a reaction of self-righteous rage to be triggered. The trigger may occur as a reaction to a real slight by another or to an unintended slight, the seriousness of which is exaggerated by narcissistic individuals’ extreme sensitivity and tendency to overinterpret rejections or disappointments. Alternatively, the trigger can be when the narcissistic individual perceives people or organizations as not living up to his or her expectations, following which the narcissistic individual classifies them as behaving badly toward him or her or even others. Celebrities or athletic stars who are no longer getting the center-stage attention they once had are examples. Such individuals then develop a sense of self-righteous morality or take up causes to reform others, both reactions that only thinly conceal their own powerseeking needs. For example, they may seek positions of authority or leadership to give them the opportunity to “correct” others. On a psychological level, what they are seeking is to rectify their feared depletion of selfesteem or inner resources before it is too late. In some way, they sense that the situation is becoming dangerous. Many civil lawsuits arise out of situations in which these individuals seek to “educate” someone they feel has depleted them. Perhaps that is a more fortunate twist than resorting to schemes of violence.

184


Developmental Factors Seen in Rage States Diverse explanations exist for the emergence of rage states. Rage states were once simply viewed as an individual’s expression of increased aggressivity, presumably temperamental or genetic in origin. Early learning theory models used a frustration-aggression model in which rage was seen as an exaggerated response to frustration.9 In terms of explanations of how some people with NPD can become homicidal, it is unfortunate that DSM-IV excludes a criterion that DSM-III-R10 had. That criterion referred to such a person reacting to criticism with rage, shame, or humiliation, and this criterion was excluded on grounds of a lack of specificity. Yet NPD individuals persistently seek power when they are thwarted, which easily fits this model. Their lack of empathy is often coupled with their lack of sensing how others are reacting to them. Other explanations shifted to view the breakthrough of rage as more connected with a breakdown of the ego defenses that ordinarily serve to control aggression.11 The episodic dyscontrol syndromes were one psychological explanation used to show the varying degrees of aggressive breakthroughs that could occur. Later, a neurophysiological model to explain rage reactions was introduced, independent of theorizing about NPD, in which the breakthroughs of rage were seen in terms of failed neural substrates of inhibition.12 The possibility of such a vulnerability in NPD and borderline personality disorder cannot be discounted in all cases. More contemporary psychodynamic theorists began to introduce concepts of disturbances in narcissistic equilibrium as an explanation for why rage reactions occur in narcissistic personalities. Kohut,13 Kernberg,14 Jacobson,15 and Mahler16 stressed instabilities of self and object representations. Developmental theorists added in the component of antecedent factors from earlier periods of strain or traumatizing experiences when a child sought certain parental responses. Kohut saw narcissistic rage as related to the child not having his or her needs met in the context of the his or her grandiosity. In contrast to the child who learns to control his or her destructive types of assertiveness, the child on a path of narcissistic development has trouble separating from key figures. In reaction to the failure of significant authority figures to be consistent in setting limits on the narcissistic child’s demands, that child develops expectations that his or her imperious needs should be gratified on demand. When those demands are not met by others, he or she experiences humiliation and as a result enters a state of being hurt or wounded. This experience of great humiliation can then lead to a more extreme state of narcissistic mortification, connected with shame and


185

the need to rid oneself of this painful emotional state. In situations of excessive indulgence or traumatization, the child is set up to believe that he or she can actually manipulate control of a situation, despite quite the opposite being true. It should be noted that Kernberg17 did not view the aggression in narcissistic people as arising solely from others inducing frustration but rather from intrapsychic origins. A deep envy was present as part of the clinical picture of wanting to destroy someone. This mistaken belief that one can control all situations could also be interpreted as a type of cognitive distortion. Children caught in such a dilemma begin to feel that their well-being is contingent on maintaining their sense of personal omnipotence and control. In some situations, the child’s grandiosity has been stimulated by excessive praise, whereas other children have been deprived of the praise they needed and developed an overcompensated self-worth. To react otherwise than by attempts to control would allow old anxieties, aroused when the child’s wishes were not met, to reappear. As a result, the child may attempt various types of distorted self-assurance. If a parental figure is not empathic to the child’s demands, the experience of helplessness sets the stage for a further development of rage reactions. When a child is forced to face the reality that he or she is not actually omnipotent, given the child’s need for control, the child is then set up to experience humiliation. The imposed need for the child to accept his or her dependence on others for maintenance of self-esteem leaves that child in a precarious balance. Groping for security, but hating dependency, the child is induced into a traumatized vulnerability in which he or she feels a mixture of shame and weakness from seeking attachments. Children who are so conflicted may then develop the potential for experiencing anxiety states and possibly rage reactions. Some children react by becoming withdrawn and apathetic, seeming to develop the descriptive characteristics of a depressed child. Others begin to develop the nucleus for diverse expressions of pathological narcissistic development. One study noted that disruptive behavior disorders diagnosed in childhood (conduct disorders or oppositional defiant disorders) increased the risk for a later NPD in adulthood by a sixfold figure.18 Adult psychopathology always looms as a potential consequence of the vulnerability in these individuals’ deficient self-esteem system. When they encounter similar stressful situations in which their control is weakened, the potential for rage reactions with accompanying violence occurs. Depressive episodes, panic attacks, somatization disorders, and depersonalized states are what cloak the violent potential. By the time these individuals reach adulthood, the unrealistic attitudes and expectations that others will be at their service have become ensconced. Maturity in years does not bring an acceptance about the un-

186


reality of their expectations. Although at times family members may gratify them, they are left feeling vulnerable, weak, or a failure. The sense of failure to themselves and their loved ones intermingles with assertions of grandiose claims and expectations about their prospects and ambitions. For the lucky ones, good relationships may have developed despite their vulnerabilities and may have altered their life course so that they become able to develop better coping skills in adulthood. Because some narcissistic individuals have genuine talents, they may have experienced situations that were genuinely rewarding despite the fact that they arose out of their manipulation or control. Having to work with others always raises the possibility of difficulties related to their grandiose needs. Yet other NPD individuals become entangled in a type of pathological development that leads to their perceiving others as either all good or all bad as part of a splitting phenomenon. Their corresponding self-image begins to alternate between being either worthwhile or worthless. At that stage, they have the tendency to externalize and view their own bad attributes as existing in the world of others. This externalization in turn allows persons with NPD to rationalize using others, sometimes leading to aggressive, if not sadistic, behavior. If a rage eruption threatens or a violent act occurs, it is seen as at least partially justified on the basis of their having projected a sense of badness onto others.

Self Psychology Perspective on the Development of Rage To understand the role of narcissistic rage, I turn to clinical contributions from self psychology and ego functioning, especially with regard to how one’s sense of the cohesion of the self emerges. Also important are the process of idealization in personal development and a child’s history of relationships with parental figures, especially as related to power relationships. Some parents relate to a child in an empathic manner to an extended degree so that a narcissistic merger occurs between them. Part of the personality of the child is then viewed as part of the parent’s own; at other times, the parents may respond to the child as having a sense of self independent of them. Within this simplified framework, all degrees and varieties are possible. In the realm of thwarted narcissistic development, consideration needs to be given to such factors as the impact of unmet ambitions on individuals’ personality functioning, reactions to those who thwart them, wishes to dominate that cannot be fulfilled, desires to be outstanding and the center of attention, and yearnings to merge with powerful figures with whom they can identify. To the extent that a person does not have a sense of ego fulfillment in one or more of these areas,


187

that person is left vulnerable when occasions arise that require mechanisms to deal with new disappointments. The individual’s striving for a total justification and triumph over another who is perceived as having damaged his or her self-esteem sets the stage for an eruption of rage. In the course of their lives, most people experience and then move on from traumas or insults to the ego, a process that requires periods of transition and new adaptations. These traumas or insults most likely reactivate emotional experiences from key earlier periods when the self was being formed, and the individual builds on his or her previous reactions to similar situations. The individual’s current reaction involves not only his or her self-esteem but also others with whom he or she is involved in relationships. In addition, it is clear that a particular environmental concatenation plays an equal role in developing those tendencies along with a damaged sense of self. A sense of the person’s vulnerability can be assessed by how he or she has handled major changes in the past (e.g., the demands of moving from one geographic area to another, separation and divorce, deaths, entering military service, changing jobs, dealing with sudden financial reverses, not getting a promotion). Such experiences provide in vivo tests for an individual’s specific weaknesses and strengths in terms of his or her sense of self, allowing a tentative model to be formed that illustrates vulnerabilities or sources of security that exist in the person’s narcissistic realm and that may be manifested in fluctuations in selfesteem. A parallel development is that throughout life, an individual’s self-representation changes over time. In summary, extensive changes in the self are required to deal with certain kinds of developmental transitions from early childhood onward, as well as the various changes in the self required in adulthood. The danger in responding to situations calling for change is that if earlier types of self-representation have remained conflicted or faulty, the rebuilding process from the new situation can be difficult. The person’s vulnerability lies in his or her preexisting type of psychopathology, which is reactivated in the contemporary context. Particularly striking are earlier conflicts involving individuation that have not been resolved.

How Narcissistic Rage Can Lead to Homicidal Behavior What specifically leads to a homicidal act in the context of narcissistic rage? How is the element to be classified in terms of clinical knowledge and dynamic factors for an individual? Narcissistic components are present more frequently in homicides than is usually believed. For example, many of the killings that are en-

188


compassed under the clinical heading of rage reaction or the legal heading of temporary insanity have been perpetrated by someone with narcissistic disturbances. A more detailed clinical scrutiny would reveal these killings as situations in which a narcissistic injury or injuries occurred and in which the killer saw a need to right what was perceived as a wrong perpetrated on him or her. The phrase “crime of passion” is an example of a type of narcissistic killing in which a former lover or spouse has rejected the perpetrator. Those who have done the rejecting, or the new mate on the scene, are killed in the perpetrator’s attempt to regain a narcissistic balance and to be in control. Although the act might seem to be precipitous, most likely it in fact developed over an extended period of time, with the killer brooding and fantasizing over the need to do something to rectify the situation, even though his or her thoughts might not actually have reached homicidal proportions. As noted earlier, the intrusion of paranoid ideation can further confuse the situation and miss the correct diagnosis of the individual’s motivations. All factors may then finally converge toward a homicidal act to resolve the person’s longstanding sense of injury. The clinical connection is ultimately between the act of violent aggression that occurs and the antecedent narcissistic vulnerability. In this context, legal complexities are added when concepts such as premeditation or intentionality are raised in courtrooms in attempts to assess responsibility for these killings. It may be very difficult to convey these developmental vulnerabilities even to a sophisticated legal audience, let alone a lay jury. Several variables lead to a final common pathway for an individual to exhibit rage reactions of homicidal proportions. The key seems to be an experience of some type of threat to one’s vulnerable self that becomes magnified far beyond the objective nature of the threat. When the person perceives himself or herself as being under threat or having the potential for becoming victimized, someone in the external world is customarily singled out as the hostile aggressor. The aggressor might actually be a person or a situation that develops because of the narcissistic individual’s sensitivity to innocuous comments, slights, or nonachievement. In response to the perceived aggression, the narcissistic individual experiences not simply fear but eventually rage if the situation is not righted. The rage may also be felt within the context of a role reversal in which the narcissistic individual’s sense of badness is externalized; therefore, when the evil is seen as coming from the external world, an attack on the bad object seems much more justifiable. If this externalization fails, a painful inner state of emptiness and depression looms. The person with an extreme narcissistic vulnerability lacks the capacity to assess the situation


189

accurately and correctly assign blame for his or her pain. Something in the cognitive processes of assessing information realistically about others and oneself is deficient. Even if assessments are temporarily accurate, they do not remain stable over time. This difficulty in processing information to assign blame is the crucial weakness. Instead, ego deficiencies set up narcissistic individuals to attack someone impulsively in a burst of self-righteous rage. Feeling humiliated, weak, and deceived by others, narcissistic individuals fall back on needing to attribute blame for someone having deliberately done things to them. They are left vulnerable to react with miscarried attempts to remedy a chronically devalued sense of self. A distinction has been made between “thick-skinned” narcissistic persons, who are grandiose and derogatory, and those who are “thin-skinned” and preoccupied, self-loathing, and more vulnerable.19 Dangerous periods arise when a shift occurs from one of these states to another. During these shifts, a lowering in affect regulation can overflow into a feeling that the narcissistic person needs to take some action and do something. The thinskinned narcissistic person’s dependency leads to an intolerable vulnerability, which might lead to an angry attack on someone seen as autonomous and who seems to be mocking his or her helplessness.

Feelings of Shame When narcissistically vulnerable individuals are put in extreme positions in which their self-esteem is severely threatened, they seem to respond in two ways: via a shame mechanism that leads to withdrawal and hiding or via an outburst of rage that leads to some type of aggressive action. In this sense, shame and rage are classified as manifestations of a disturbance in narcissistic equilibrium. Some clinicians portray these individuals as being caught in a vicious repetitive cycle, with increased anger leading to guilt, guilt leading to submissive types of behavior, submission leading to reactive aggression, aggression leading to more guilt, and so on. The failure to individuate from those with whom the person has close attachments, as seen in the following case example, is frequently found.

Case Example A 19-year-old male, who had once been a fine student, repeatedly landed himself in difficulties because of his drinking and marijuana use. During his high school years, countless people became involved with him in an effort to “get him back on track.” Clinics, inpatient hospital units, and counseling were all tried. In the year after graduation from high school, he became insistent about his need to get away from home and his overcontrolling parents. However, after a period had passed

190


without his taking any action, others became indifferent to his complaints. Instead of remaining sympathetic they would simply tell him, “You’re 19 years old. Go ahead and leave.” His forays of leaving always resulted in excuses to return and a failed sense of being able to separate. An attempt to get the young man involved in therapy ended when insurance coverage for the therapy expired. Eventually a major argument occurred between him and his parents, who began to tell him he was no longer welcome to stay at home. After having words with his father, the two physically fought. At one point the son struggled loose and took a brief walk. Feeling shame and humiliation for having hit his father, but more generally for having failed his parents over several years, he returned home, went to his room, and got a gun. Pausing outside the room his parents were in, he heard them agreeing he would have to leave. He felt further ostracized, entered the room, and shot them both.

The following is another case in which perceived slights and humiliation led to a homicidal outcome.

Case Example A man did not receive a promotion in his job that he thought he deserved. Being a compulsively dedicated employee, he had long assumed that hard work and dedication, such as spending far more hours in the office than required, would lead to a promotion. In many ways his situation mirrored what has been referred to as the “Nobel Prize complex,” in which the person’s grandiosity has originally been fed and the person believes that he or she is headed for greatness.20 What the man was not registering was his lack of interpersonal relationships with many of his colleagues. The applicant who did receive the promotion came from another division but had once worked under him. The next affront to his narcissistic tranquility came when he did not get a salary increase. He rationalized that it was because of a faltering economy. The next year he neither received a salary increase nor a bonus and had more difficulty assimilating these affronts. He then sought an explanation from his supervisor but received only vague answers. For several weeks, he brooded that the supervisor was the person thwarting him and responsible for his depleted sense of self. This thought originated not from a sense of paranoid delusional thinking but rather in the context of his needing to find the person primarily responsible for dealing him the injury. He ignored the results of evaluations of his work performance and personality done by different people over many years. At this stage, as later revealed by his wife, his behavior began to change and he began to have outbursts of rage toward her and their son over minor transgressions or failures on their part to do things just the way he wanted them done, such as his wife being late for meeting him or his son not achieving top grades in school. He was reacting with violent anger to minor frustrations caused by people who were important narcissistically to him but who were failing him in little ways when his ego needed major infusions of support.


191

He also began treating others at work unduly harshly, as a result of the narcissistic injury that his failure to gain recognition caused him. He rationalized his behavior toward his wife and son on the basis that they were better off for his being consistent in his anger to them and that being too tolerant was not good for his son. On occasion, he would drink to excess and hit them but then blame his actions on the alcohol. He eventually found that his behavior at home created family difficulties but did not alleviate the narcissistic hurt from his employer. At that time, he began to develop anxiety-related and depressive symptoms of sleeplessness, periods of agitation, and increased hypochondriacal concern about excess flatulence and pressure in the sternum. He was unaware of the increase in his sarcastic rage. Eventually, his lack of empathy after these rageful outbursts, outbursts that he rationalized as being justified by his being morally right, led his wife to leave him. Finally, his conflict led to action. He constructed a bomb, wrapped it in a brown paper bag as if it had been mailed, and gave it to a secretary to deliver to his supervisor. When the secretary became suspicious, the package was checked and found to contain an explosive that could detonate. The discovery led to the man being investigated and criminally charged with attempted murder, after which he received psychiatric intervention.

Need for Revenge Narcissistic rage often exhibits the person’s underlying need for revenge. In clinical assessments of those who have already acted on the basis of murderous rage, it becomes clear that they were not able to move beyond the wrong that had been done to them and the need they felt to induce hurt or retaliation. In its more extreme manifestations, the need for revenge arises in the context that an individual has suffered a persistent and unresolved narcissistic injury. Dedicated efforts to right the wrong now begin. This sense of righting a wrong distinguishes narcissistic individuals’ behavior from others’ simple expressions of aggression. The trigger that stimulates this type of narcissistic rage can be quite minor in someone narcissistically vulnerable to begin with, such as an episode of ridicule, teasing, or contempt or something experienced as a defeat. Children who have been passive recipients of shameful or sadistic treatment by others (treatment that is sometimes overt, such as physical or sexual abuse, but more often covert, such as being shamed) are left with unresolved desires to gain control so that they never again have to experience such helplessness. Some researchers interpret the eventual reactions of these victims as representing the mechanism of identifying with the aggressor. Whatever the specific experiences of these individuals, they are left in a state of readiness to react when they feel they are once again being

192


placed in a shamed position. They are ready to take action to inflict on others the type of narcissistic injury they are afraid will again be imposed on them. The shame imposed on them leaves them later vulnerable to becoming enraged over the shame and to becoming homicidal. In this context, some researchers have viewed the experience of childhood physical or sexual abuse as the sine qua non for these individuals to try to avenge themselves later in life when they perceive themselves in a situation of being used. An example would be the battered woman syndrome, in which a woman has been physically abused in the context of a relationship. If that woman later murders the lover who abused her, the legal defense of duress or self-defense may be used, even though the lover might not have been attacking the woman at the time he was killed. The woman’s state at the time of the killing would correspond to that of an ongoing narcissistic injury, although the legal defense would be in terms of her living in a state of imminent attack.

Acting on the Need for Revenge Although everyone reacts to narcissistic injuries with feelings of embarrassment or anger, narcissistically vulnerable individuals react primarily with an intense shame. Apart from the painful aspects of feeling shame, a violent eruption requires one additional step: the hope that by destroying someone, they will now gain or regain a sense of mastery over an environment that has not been conducive to their self-esteem. They are caught in the contingencies of needing the unconditional availability of approval and admiration from others, or those with whom, on a psychological level, they have become fused. Within this matrix of narcissistic vulnerability some of the most violent homicidal behavior with a lack of empathy is exhibited. Experiences of being hurt and wounded can accumulate from multiple occurrences. Narcissistically vulnerable individuals then begin to think that it is not right that they alone should be made to suffer. Their thinking takes on magical qualities to the degree that they believe that only when others are removed can their lives regain their narcissistic buoyancy. Their constricted thinking is evident in their believing that they cannot attain a sense of security or tranquility until they have either harmed or destroyed a particular individual who has thwarted, disagreed with, or overtly taken opposition to them. Hence any threat to their supposed autonomy or self-sufficiency is removed. These individuals’ need to blot out some seeming offense against them, coupled with the fury from the felt lack of control over the person or object whom they think has offended them, is the final step toward a


193

homicide. That is why the events that have happened to them—often minor slights or events that occurred years earlier—are reacted to with such a disproportionate degree of severity. At the extreme, a preoccupation with revenge may lurk for years. An unceasing compulsion to square accounts with someone after some presumed or actual offense can also arise anew years later. In one case, a man who believed he was sentenced unjustly by a judge began to write anonymous threatening letters to the judge 15 years after the trial. The result of a brooding need for revenge can be not only a “clean” act of killing, such as a shooting, but often a brutal killing—such as a prolonged beating, pouring a chemical (e.g., acid) on someone’s face, dismemberment, or multiple wounds. Themes of revenge in response to narcissistic injury abound in literary works. The Count of Monte Cristo and Moby Dick are two classic examples. Kohut 21 referred to Michael Kohlhaas, a well-known work in Germanic countries, written by the great dramatist Heinrich von Kleist (1777–1810), to illustrate what he believed was a separate line of narcissistic development. The insatiable search for revenge subsequent to a narcissistic injury in Michael Kohlhaas has been interpreted differently by Hamilton,22 who believed it could be traced to unresolved oral conflicts giving rise to a primitive rage that is acted out. von Kleist actually ended his own life by suicide. The story of Kohlhaas, briefly retold below, well illustrates the dynamics that can lead to narcissistic types of killing.23 Kohlhaas was a horse dealer and small-land owner in sixteenth-century Germany. Those who knew him saw him as an honest but, at the same time, terrifying man. Until age 30 years, he was a model citizen, but a sense of injustice turned him into a robber and murderer. The event that led to his subsequent transformation was a grievous and entirely unprovoked humiliation that he received at the hands of the servants of a nobleman in his community. From the various machinations of these servants, Kohlhaas lost his property; his efforts to redress the matter legally via petitions went unanswered because the petitions were intercepted by relatives of the culprits who had wronged him. These relatives held key positions at the Court of Saxony; hence, the prince remained oblivious of the wrongs that were being done to Kohlhaas. These events led Kohlhaas to become a rebel against his sovereign. He assembled a gang and began to burn and pillage villages, in the course of which inhabitants were robbed and killed. Martin Luther issued an appeal for him to come to his senses, pointing out that the injustice had been done by subordinates and that the sovereign had no part in it. Kohlhaas responded to Luther, who then intervened with the sovereign for Kohlhaas to have a safe passage to get a hearing before a competent tribunal. Luther also asked for a pardon for Kohlhaas’s rebellion.

194


A hearing was held, but court intrigues continued. In the end, the tribunal granted Kohlhaas full compensation for the losses he had suffered for his property and punished his enemies. Here, an ordinary story would end. However, the court also felt justice would be served only by Kohlhaas being punished for his crimes and ordered that he be executed. Kohlhaas willingly and joyfully accepted his punishment because he had achieved satisfaction for the injustices done to him.

The Kohlhaas case is often held up as an example of the perfect blending of legal and moral justice for a lawbreaker. From a psychiatric perspective, the case can be conceptualized as killing from a sense of injustice. At times, the injustice could be framed as a social injustice, without looking at internalized conflicts. However, this sense of injustice can also be correlated with internalized broodings and unresolved conflicts from perceived wrongs, sometimes from decades earlier, in which homicide is viewed as an ultimate act of resolution. Such social and personal struggles are a reminder of an argument by the great jurist Oliver Wendell Holmes Jr.24 that legal systems have their origin out of a need for revenge. The need for revenge and retaliation elicits retribution as the major goal of a criminal justice system. To achieve this goal means that anger and rage must not be allowed to fade, because they are the driving force to achieve revenge. Societies in which honor is taken seriously, such as in the American South or in the inner cities, are thus more violent than cultures in which honor is taken less seriously. When dishonor or disrespect becomes intensified and leads to preoccupation, a touchiness is created in which slights are sensed even when none are intended.25 Thus, a perception of disrespect plays a vital role in a sense of injustice and seeking retaliation.26 In this context, the pursuit of justice has assumed a prominent role in American society. The philosopher John Rawls27 pointed out that justice is not only a concern in a person’s everyday life but the first virtue of social institutions.

Dissociating From the Act of Violence The potential is ever-present for violent acts to escape integration with the remainder of the personality. Aggression is then expressed within the framework of some primitive personality reactions and misperceptions of reality. The primary deficit of an exaggerated assessment of reality plays a key role in an ultimate homicide. This deficit is not the same as the pervasive dismantling of reality testing that occurs in a psychotic reaction. Rather, the deficit is that the narcissistic individual’s reactions are shamed-based, the person having experienced reversals or disappointments as narcissistic blows to a degree that he or she has little


195

resilience left. A state of readiness to respond with an outburst of rage toward those people who are in close relationship to the individual or a displacement to more distant figures is a possibility. Those people the narcissistic person attacks are seen as having thwarted or offended him or her rather than being seen simply as independent individuals with different purposes or behaviors. In a developmental sense, the adult with this type of archaic personality structure has a cognitive set that remains isolated from the rest of the personality. From a different perspective, expressions of narcissistic rage can be seen as enslaving the ego and allowing the individual to function only as the tool and rationalizer of a deep hurt that has never healed. When a homicide is perpetrated by a person with NPD, others are often surprised by the lack of remorse in the perpetrator. Part of the explanation is that the victim is not viewed by the perpetrator as a human being with his or her own needs and feelings. Rather, the victim is simply viewed as having been a drain on the perpetrator’s life or a persistent source of his or her own suffering. As such, there is no empathy with the person destroyed. The deficit in empathy with the victim, as well as the victim’s relatives, is witnessed in the self-justification for the acts. In turn, the perpetrator’s response is perceived as that of not taking responsibility for his or her actions, either in a legal context or a therapeutic setting.28 In the narcissistic person’s mind, the need was to eliminate the source impinging on his or her failed need for control. The individual had hoped to restore his or her grandiosity and avoid further limitations. However, the reality of a homicidal act has intervened. Although some, such as Kohlhaas, may feel that the psychological victory was worth the price, not all do. For most, their victory is celebrated by a long period of incarceration in either a prison or hospital facility. The lack of remorse often experienced by individuals with NPD is illustrated in the following case example.

Case Example A young woman was told by her boyfriend that he no longer wished to continue their relationship. She asked him to continue contact through phone calls. In a few weeks, he told her he no longer even wanted to do this, because the relationship had become uncomfortable for him to maintain. However, she continued to call him at home at various hours of the night and at his office in the daytime, but he never responded to these calls. He felt that the relationship was over and began to pursue a new social life with a sense of relief. At the time, he could not verbalize why he felt such relief, beyond noting that he was enjoying a new social life. In the course of time, he began to date several other women, occasionally frequenting places with them that he had gone to with the

196


former girlfriend. After several months, on the pretext of needing to see him about some serious matter, the former girlfriend arrived at his apartment. At a time when his back was turned, she fired several shots into his back, killing him. She expressed little subsequent emotion but instead displayed an apathetic type of depression with no remorse and minimal anxiety.

PSYCHODYNAMICS OF NARCISSISTIC KILLING When these types of killings are examined in more detail, the operation of homicidal aggression rooted in the matrix of an archaic type of narcissism becomes clear. Threats and blows to this narcissism can give vent to one or more acts related to the rage engendered. In this context, it would be erroneous to assume that the homicidal behavior always occurs in the form of wild or out-of-control outbursts. The behavior may take the form of orderly and organized activities to secure revenge for felt injustices, in which the homicidal act is carried out with an absolute conviction about its righteousness. In fact, acts of retaliation may have been held under control for some time by way of an enduring sense of hatred that functions as a homeostatic adaptation.29 In some cases, a person may feel shame and withdraw in response to being humiliated, with these initial reactions setting the stage for a later overt enactment of rage. What persists is the need for revenge, to right a hurt that may have been simmering for some time. For some individuals, the need simmers and remains dormant and later becomes activated by some precipitant. In other cases, the unresolved hurt consumes an individual over a lifetime, with a compulsive intensity that gives no surcease. It is important to analyze the dynamics to determine what finally elicits a homicidal level of violence in these narcissistic personalities. An initial question is to assess whatever external precipitants may have left residues that have not dissipated over time within a particular person. More immediate events that have been operating, such as provocative acts or words, shaming techniques used once too often, bullying behavior, ridicule, or lording something over the person, are a few possibilities. The level of sensitivity within a rage-prone personality also needs appraisal. Such approaches usually lead to the uncovering of shame-based reactivities within the person, which leave him or her particularly ready to strike back or to ward off anticipated attacks from others. Note that given such background factors, narcissistic persons may have some type of plan or devious course of action ready to set in motion, but they also may be oblivious to the risks they are taking and to what the ultimate consequences can be to themselves.


197

The dedication to a course of action is similar to that witnessed in individuals with borderline personality disorder or even delusional disorder. The former diagnosis is sometimes only distinguished with great difficulty and fine distinctions. The latter diagnosis has the hallmark of beliefs not subject to reality factors, even though the semblance of logic is present. Often, NPD is differentiated by the sway of a powerful need for righting a perceived wrong and a sense of personal injustice. It is no exaggeration to say that to eliminate a sense of hopeless mortification, these persons are sometimes willing to pay the price of destroying themselves. This perspective is similar to the psychology of fanaticism that can be enlisted for any cause. Besides righting what is experienced as a wrong inflicted on the self, these acts are perceived to eliminate shame from having tolerated such past behavior toward oneself. In that sense, the imperfections of the self gnaw at the person, which elicits a secondary rage. The sense that the individual cannot control his or her environment, or his or her life, is now to be righted. Diminished selfesteem is to be cured by a desperate effort to gain control.

REFERENCES 1. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 2. Gunderson JG, Ronningstam E: Differentiating narcissistic and antisocial personality disorders. J Personal Disord 15:103–109, 2001 3. Stewart JB: Den of Thieves. New York, Simon & Schuster, 1991 4. Gabbard GO: Psychodynamic Psychiatry in Clinical Practice, 4th Edition. Washington, DC, American Psychiatric Publishing, 2005 5. Garfield D, Havens L: Paranoid phenomena and pathological narcissism. Am J Psychother 45:160–172, 1991 6. Coid J: Correctional populations: criminal careers and recidivism, in Textbook of Personality Disorders. Edited by Oldham JM, Skodol AE, Bender DS. Washington, DC, American Psychiatric Publishing, 2005, pp 579–606 7. Horowitz MJ: Formulation of states of mind in psychotherapy. Am J Psychother 42:514–520, 1988 8. Warren JI, Burnette M, South SC, et al: Personality disorders and violence among female prison inmates. J Am Acad Psychiatry Law 30:502–509, 2002 9. Berkowitz L: The frustration-aggression hypothesis: an examination and reformulation. Psychol Bull 106:59–73, 1989 10. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987

198


11. Lewis M: The development of anger and rage, in Rage, Power, and Aggression. Edited by Glick RA, Roose SP. New Haven, CT, Yale University Press, 1993, pp 148–168 12. Monroe RR: Episodic Behavioral Disorders. Cambridge, MA, Harvard University Press, 1970 13. Kohut H: The Restoration of the Self. New York, International Universities Press, 1977 14. Kernberg OF: Pathological narcissism and narcissistic personality disorder, in Disorders of Narcissism: Diagnostic, Clinical, and Empirical Implications. Edited by Ronningstam EF. Washington, DC, American Psychiatric Press, 1998, pp 29–51 15. Jacobson E: The Self and the Object World. New York, International Universities Press, 1964 16. Mahler MS: On Human Symbiosis and the Vicissitudes of Individuation. New York, International Universities Press, 1968 17. Kernberg OF: The psychopathology of hatred, in Rage, Power, and Aggression. Edited by Glick RA, Roose SP. New Haven, CT, Yale University Press, 1993, pp 61–79 18. Kasen S, Cohen P, Skodol AE, et al: Childhood depression and adult personality disorder: alternative pathways of continuity. Arch Gen Psychiatry 58:231–236, 2001 19. Bateman AW: Thick- and thin-skinned organisations and enactment in borderline and narcissistic disorders. Int J Psychoanal 79:13–25, 1998 20. Tartakoff H: The normal personality in our culture and the Nobel Prize complex, in Psychoanalysis: A General Psychology. Edited by Lowenstein RM, Newman LM, Schur M, et al. New York, International Universities Press, 1966, pp 222–252 21. Kohut H: Thoughts on narcissism and narcissistic rage. Psychoanal Study Child 27:360–400, 1972 22. Hamilton JW: Implications for current concepts of narcissism in Heinrich von Kleist’s “Michael Kohlhaas.” Int Rev Psychoanal 8:81–84, 1981 23. von Kleist H: Michael Kohlhaas (1808). Edited by Geary J. Oxford, England, Oxford University Press, 1967 24. Holmes OW Jr: The Common Law. Boston, MA, Little, Brown, 1881 25. Posner RA: Law and Literature. Cambridge, MA, Harvard University Press, 1998 26. Miller DT: Disrespect and the experience of injustice. Annu Rev Psychol 52:527–553, 2001 27. Rawls J: A Theory of Justice. Cambridge, MA, Harvard University Press, 1971 28. Paris J: Personality Disorders Over Time: Precursors, Course, and Outcome. Washington, DC, American Psychiatric Publishing, 2003 29. Galdston R: The longest pleasure: a psychoanalytic study of hatred. Int J Psychoanal 68:371–378, 1987

CHAPTER

7 MASOCHISM AND HOMICIDE The Ultimate Enslavement

THE COMBINATION OF victimology, battered partners, and masochism with the subject of homicide may at first seem an unlikely one. Yet when taken together, the first three phenomena represent a prime mixture from which a homicide may eventuate, the complexities of which may never be resolved. These issues are often raised in courtrooms in cases in which spouses or lovers have become enmeshed in distorted processes of intimate relationships. They may also arise in the complexities of love, hate, guilt, provocation, and punishment that go on in sadomasochistic interactions that can end in homicide. Some of these are personal interactions that get distorted by power battles ending in a homicide. Others may be related to sadomasochistic fantasies initially used to try to deal with anger and fears of being attacked by others, but at some point the zone of protection or safety is gone. The transition from fantasized sadomasochistic actions, sometimes safely enacted with a person, to coercive acts is what can lead to a homicide.

VICTIMOLOGY AND HOMICIDE Defining the Victims of Homicides Apart from one obvious answer—that by definition, the victim of a homicide is whoever is killed—more subtle answers are uncovered when the events leading to and following a homicide are examined. For ex-

199

200


ample, those indirectly affected by the killings—children, parents, siblings, friends, lovers, and so on—also must cope and deal with an event that will have a major ongoing impact on their lives. Victimology is a theme running through many cases of individuals who perpetrate some type of homicidal act. Indeed, an approach could be taken that views all homicides solely from the perspective of the perpetrator being the victim. Such a perspective would be particularly useful with various types of personality disorder diagnoses. Yet questions can then be raised about who, in fact, is to be labeled the perpetratorvictim in such interactional situations. Among those who should be considered victims are people whose disrupted lives and relationships may have led to the homicide. Another approach could be to view all perpetrators as being victims. The perpetrators’ victim role customarily involves spending decades incarcerated in a prison or a security hospital, if the perpetrator is not executed. The issue pertains to the perpetrators’ lack of understanding of the diverse ways in which they have unconsciously pursued such a sacrificial ending throughout their lives. In the end, there are only a dead victim and multiple other living victims.

Epidemiological View There are two key questions to answer in terms of victims: 1) Who, in a certain group, is susceptible to being the victim of a potential act of homicide? and 2) What makes that person susceptible to such a lethal outcome? An epidemiological approach would be one way to try to detect a group of those most vulnerable. Such an approach reveals homicide perpetrators and victims clustered with a few key variables. In Chapter 1 (“Epidemiological Aspects of Homicide”), I noted that one group of those most likely to be involved overall in homicides is predominantly young males, especially young black males in large urban centers or in impoverished inner cities. Other groups would be women caught up in painful emotional entanglements, elderly women, or children. Yet greater specificity is needed in each grouping, such as why do young males perpetrate these killings more frequently in certain areas of the country, or what are the psychological and social variables that operate in the lives of these women? Questions are raised about whether an increased number of women in the workforce, with a resulting increased exposure to dangerous situations and people, increases their homicidal risk. (However, caution must be taken to ensure that this hypothesis does not lead to the oversimplified assumption that a person’s increased chance of being a victim of homicidal violence can simply be correlated with his or her increased exposure.) These types of questions

Masochism and Homicide

201

are geared toward examining what is involved in the early stages leading to homicidal violence, even when the parties involved are not registering the potential situations they are living in.

Legal View: Only One Victim In the current era of victimology, discussions about the nature and sources of victimology are easily confused. Distinctions need to be made for heuristic purposes as to how the word victimology is being used in discussions about killings. From a legal perspective, the focus is on the person killed as the victim. Efforts are then directed at finding the person who can be charged with the act so that blame can be assessed. Blameworthiness is the key concept within the traditional confines of criminal law. An effort is then made either to punish the perpetrator in a retributory manner or to punish him or her as a deterrence to others from performing such an act. Although punishment might also be justified from the perspective that it might deter the same individual from again performing such an act, this emphasis may be passé given the long-term consequences of a conviction whereby the person is removed from society for an extended period. In some situations, incarceration of individuals who have killed another may be predicated on the need to incapacitate them because they are viewed as, or assessed to be, dangerous and therefore must be confined. In the not too distant past, the possibilities of rehabilitation were frequently cited for incarcerated criminals, but this goal now seems sadly antiquated for those who commit homicide, given the current mores and emphasis in the American criminal justice system on punishment and long prison terms— and the way legislatures either create public opinion or react to it. In discussions about the diverse purposes of criminal law and homicide, whatever ideas about victimology surface exert a covert influence in terms of assessing responsibility for acts and degrees of punishment. The difficulty with a narrow legal assessment of who is seen as a victim and what is to be done to the killer-victimizer is that it bypasses the major part of the criminological and clinical work that has been produced about victimology. Sociological approaches have zeroed in on the relationship between the victim and the killer, focusing on a host of demographic variables about each, such as their ages, socioeconomic status, family structures, and personal histories of violence, among other characteristics. Clinical approaches have inquired into prehomicidal behavior that may have existed and interactions between the perpetrator and others that may have led to a homicide. A psychiatric approach would also address clinical issues involving diagnosis and the intrapsychic and interpersonal dynamics associated with a particular killing.

202


The Process of Becoming a Victim Homicidology, like suicidology, involves a reconstruction of events after the fact. As more knowledge is gained, more indicators become available about what may have been premonitory signals. However, current knowledge is focused on the early stages and the natural course of events leading to homicide.

Perception of Self as Victim When homicides are studied retrospectively, it is clear that many eventual victims did not perceive themselves as victims at all in the early stages of events leading to a violent outcome, even though they were caught up in a potentially explosive situation. Their own predisposing personality conflicts and reliance on denial and avoidance may have prevented such awareness. An equal contributor could be traditional or cultural values that were operative in reinforcing their defenses against awareness of being in this role. These victims may have seen themselves as being in a situation no different from that of many other people around them or from what they had witnessed as part of their earlier lives. The masochistic tendency to attribute responsibility to themselves for what was occurring is often the most difficult part of their conflicts, because this tendency blinded them from perceiving themselves as entitled to a less conflict-ridden life and a more worthwhile existence. For another group, a quality of self-righteousness operates that drives them into rigid and unbending moral positions that can lead to victimhood. One pattern often seen is that future victims continually return to crisis situations, the same type of relationships, and the same milieu in which they have been previously degraded or harmed. A prime example is those who are involved in battering marital or love relationships. This observation does not negate the value of crisis intervention, which extricates a person from a highly volatile or dangerous situation, but for purposes of understanding as well as future prevention, it is necessary to grasp the patterns of complexities that exist in such relationships. A transitional change occurs when individuals begin to sense they might conceivably be a candidate for some future violent act, even to the point of wondering if they might be killed. To understand this process in an individual means a shift from an epidemiological approach of statistical possibilities to the psychological factors operating in a particular individual who is a participant in such processes. One of the major difficulties in assessing the process is that the role of victim is present in many different individuals and under diverse conditions. In addi-


203

tion, the term victim is now so widely applied that unfortunately it is used to ward off any critical analysis, in which attempts to understand the victim are made to seem tantamount to blaming the victim. Hence discussions of the concept become confused. In the past, the opposite situation existed, with roles of victims being ignored or tolerated by a society that simply viewed victims as living out their life destiny in this manner. The question of why some people experience extreme forms of violence, even to the degree to which their lives have been threatened, yet do little in response has been debated at length. Explanations will continue to be offered that vary from the “culturally trapped” to the “masochistically bent” model. The question of why some individuals continue to remain in a victim role is a very complex one, and sometimes, misguided efforts by “rescuers” to extricate these individuals only increase the potential for the rescuers being either a victim or a perpetrator of homicidal violence. At the boundary between psychiatric theories and theories of social psychology, the latter viewing victimization as acceptance of a role, lies the realization that the attitude and acceptance of being a victim is not primarily a conscious choice. More likely, becoming a victim results from a confluence of a set of beliefs and attitudes conveyed to certain individuals about what expectations they are entitled to in the world. The question also involves victims’ ideas about their capability of using their initiative and not necessarily seeing themselves as responsible for those who treat them poorly. In some cases, the miscarried steps of being a rescuer of others lead to becoming a homicide victim when the rescuer is not able to extricate himself or herself from a relationship where he or she is abused but believes he or she must continue to minister to the abuser. In addition, there is often a lack of viable options in the social world of some of these people. For these people, the easier choice at the time may seem to be simply adapting to a role than risk the consequences of protest or resistance. Such an adaptation is especially easier for younger people. However, all too frequently, it is easier to see oneself as the victim of a system that is “doing one in” than to take the difficult course of protesting and groping one’s way out. In a critical analysis of victims, situations must be distinguished between one extreme, in which a person sees all his or her unhappiness and misery as something unjust for which others are responsible, and the other extreme, in which circumstances are all accepted as his or her due. When a person reaches a stage in which the transition into being a victim of extreme acts of violence has occurred, prying the person from that role is quite difficult. The need to feel in control of what happens in

204


one’s life can lead to desperate measures to avoid the alternative state of helplessness. More-extreme traits of a given personality disorder then emerge, and defenses become mobilized. The experiential state of seeming helpless, without any way out of a dilemma, can give rise to contemplating desperate measures. For example, people with a pathological dependency may easily have their feelings of being abandoned aroused in the matrix of hopeless entanglements in love relationships that are seen as having no apparent exit except by a major act or declaration. What they do not comprehend is the degree of their destructive rage and hatred toward the object on which they are dependent.1 The situation is a coercive demand to be cared for and to remain in a relationship, but it is a time bomb if the entitlements that go with one’s being exploited are no longer operating or are threatened.

Reacting to One’s Sensed Helplessness Those caught in the victim role have several choices in how they can react to their role. Most of these individuals, with or without therapeutic intervention, somehow confront their life situations and try to initiate actions for change; they do not see the possible solutions to their dilemmas as being fixed in some deterministic manner. They cast off their victim role with its accompanying rationalizations and confront the situation, which they previously viewed as being inescapable other than by ridding themselves of their tormentor in some final solution. Only a small minority of individuals actually come to an end-point where they see no solution for their dilemma except through an act of killing. The Menendez case in California is a classic example, in which two brothers killed their parents, claiming earlier childhood sexual molestation by the father.2 (The increasing popularity of this type of defense merits discussion. In this defense, antecedent abuse is claimed as a justification for a homicide, with the added basis of fear that the perpetrator’s own life was in jeopardy. There is now increased skepticism of the role of antecedent abuse in homicide, in view of the widespread introduction of this abuse as a partial defense to a homicide or as a factor to mitigate sentencing.) The decision that the individual makes is then added to the many variables operative in determining the final outcome of the situation. A typical initial option is to report the behavior to police, a social agency, or a center for victims. Even when these types of intervention are successful, at least in terms of removing the person from the victimizing situation, the potential for danger can rise even further because the person is ultimately disappointed by this result; often it is their hope that only the abuse, and not the relationship, will end.


205

In one variation, one or both parties in the relationship seek therapeutic intervention. Here, too, more variables—in terms of the quality of the intervention and its aftermath—are added into the equation leading toward a final outcome. Although almost all providers refer to their programs as therapy, vast differences exist among them. Some advocate working around a high-risk relationship without trying to terminate it, whereas others advocate strengthening a person’s adaptive techniques in the short run, with an ultimate goal of terminating a relationship. Practical questions abound: How far should therapeutic interventions go to make a person uncomfortable with a role he or she is in? Might such interventions add to the stress a person is already experiencing and thereby indirectly lead to violence by such efforts? Perhaps more than in any other field, the distinctions among therapeutic intervention, advocacy, and actually taking over the decisionmaking functions for a victim become blurred. The consequence of disrupting existing relationships through therapeutic intervention usually affects several people, especially if legal action against some type of earlier violence is brought as a consequence of the therapy. The participants have an added stress of dealing with official agencies or court systems, with the attendant lack of privacy and possible publicity. The process of going public with one’s plight as a victim of violence or abuse (e.g., reporting abuse to the authorities, publicly separating from the accused perpetrator) in itself may carry the label of being a “loser.” Multiple factors operate to entrap people in abusive relationships, such as the degree of commitment to the relationship (e.g., presence of children), poor-quality alternatives to leaving, and the nature of their interdependence.3 Even if a person is an innocent victim, this label raises connotations of being viewed as someone not in charge of his or her life, unable to personally deal with his or her life situation. A connotation of incompetency may be why a male may find it difficult to admit to being a victim. The situation is complicated further when the victim is urged to go public by a counselor or someone whom he or she has consulted. In addition to the major social consequences of such an action, the individual may also have the accompanying realization that he or she did not initiate the action but was simply carrying out someone else’s (e.g., the counselor’s) plan—which in effect maintains the person in a helpless position. In many cases, the individual is not apprised of the ramifications of going public and experiences more anger and desperation. Currently, a great effort is made to avoid blaming the victim for his or her predicament. Yet such an effort alone cannot negate the subsequent fallout from the helplessness experienced by the victim. This con-

206


sequence is even more acute for the adult victim of serious violence than for a child or elderly victim of abuse. The reason is that a child or elderly person is viewed differently than is an adult involved in a relationship with his or her peer (i.e., another adult). The quality of shame in being a potential victimizer-murderer, as well as a potential victimmurderee, leaves both sides periodically wanting to expunge this part of their lives and start over. Yet their past leaves them vulnerable to extortion to conceal the situation. The extortion can be literal in some cases and psychological in others. By keeping past outbreaks of violence private, the person is hoping, or betting, that homicide will not result. (Based on statistics, the probability of such a violent outcome between partners is small.) In many situations, however, this privacy is only relative. Often, individuals being victimized informally share their dilemma with friends or neighbors. Then, given the number of individuals who are involved in any type of professional or educational role, in which case they would be required by mandated-reporting legislation to report violence, it is very difficult to keep such family situations private. These mandates apply not only to legal minors but also to those who are interpreted as being in a “vulnerable adult” status. Also, dutyto-warn statutes or case laws state that certain professionals are to warn potential victims of any violence, presumably so that potential victims can act to protect themselves from possible danger (“duty to protect”). Such legislation and legal cases put the burden on professionals to carry out this official and public reporting. Again, although these laws and statutes may thus protect some potential victims, it is not clear how often they do, and presumably the professional relationship with the patient has been changed.

Sociological View on Becoming a Victim Some emphasize that an individual is simply trapped in a social system that produces victims and will continue to do so. Hence, unless there are societal changes in the underpinnings of this system of victims, violence can be expected. A frequent example is that of women caught up in a patriarchal family system with partners who view them as their property to use or abuse as they wish. The victims of such abuse are viewed as having little chance of altering the situation because they are seen as fighting a cultural system. A less drastic view is that such individuals are burdened by trying to function in such a system. However, the system is not seen as monolithic but rather as a system in the process of change. More radically opposed views do not necessarily see institutionally fixed mores being as controlling as they once were. Although


207

the significance of personality characteristics is denied by some in favor of a cultural determinism, these characteristics operate even if one concedes that cultural factors exert a great influence.

ABUSED PERSONS AND HOMICIDE Increasingly, cases in which a person who was once abused strikes out against his or her abuser are being seen in the courts, with the battered spouse syndrome and its variants being raised as a legal defense. Only the dramatic cases of this phenomenon, such as a wife cutting off the penis of her husband after he has beaten her or children killing parents after being abused by them, receive constant media coverage; however, more mundane versions of these cases are played out daily in courtrooms across America. Because of the increasing use of this syndrome as a defense for homicide, this type of homicidal behavior needs to be examined not only from a legal perspective—that is, whether it is justified in cases where someone has been so victimized, and what responsibility that person has for those actions—but also from a clinical perspective—that is, trying to understand such homicidal behavior. From the latter perspective, it must be recognized that diagnostic categories never arise in a scientific vacuum; this principle applies both to battered spouse syndrome, which in some cases may be associated with posttraumatic stress disorder (PTSD), as well as to some other diagnoses that involve patterns of self-defeating or masochistic behaviors. Using social theory on victimology, coupled with psychiatric knowledge of personality disorders, one can obtain a broader consideration of the complex and controversial issues surrounding a battered spouse killing his or her partner. Although some of these cases involve male perpetrators, the issues are most often raised when a female commits a homicide. In this section, I focus the discussion on situations in which a woman has killed a male partner whom she alleges has physically abused her in the past. I will not look at the vagaries of verbal abuse that could allegedly lead to sufficient provocation to kill a partner, although these cases do arise in the context of a claim for killing in self-defense. More recently, cases of juveniles who commit acts of parricide have raised issues of battered child syndrome. Whether the analogy to the situation of battered women can be upheld by clinical evidence and in courtrooms remains elusive.

Who Are Battered Women? An Interpersonal Perspective Descriptions of the interpersonal lives of battered women are different from many of the stereotypes provided. A false stereotype is that the

208


battered woman is a passive person who does not fight back. In fact, most battered women have fought back at some time. This fact in itself is subject to different interpretations. One view is that their fighting back is an effort to bolster their sagging self-esteem. Another is that these women learn to express their anger and hostile feelings by striking back. In one study of 400 battered women, 92% of the women thought the batterer could kill them, but 54% also thought they could kill the batterer.4 Such data indicate the magnitude of the risk for a potential homicide within such a population. In addition, 87% of those women stated that they thought they would be likely to die during a battering incident. Threats of suicide were present in 48% of the batterers and in 36% of the women. None of the women perpetrated a killing, although most of them had fantasized about the batterer’s death in the course of their relationship. In some cases, this admission of having fantasized about the killing is raised to challenge the woman’s self-defense strategy by claiming that the killing had been premeditated. Browne5 calculated that 12% of homicides in the United States are committed by women and that most of them are committed by battered women who allege they killed in self-defense. Several points merit discussion here. The first is that these cases arise from a variety of background situations; often the only common thread among them is the women’s claims that being battered led them to the homicide to protect their lives. In some cases, there is a background of physical beatings administered by the male in the relationship, with several earlier and unsuccessful attempts by the female partner to extricate herself from the situation. At the other extreme is a man who strikes a woman during a heated quarrel, which culminates in her striking back and killing him. In yet other cases the killing occurs hours or days after an argument. Sometimes the homicide is planned far ahead of time, being seen as “the only way out,” and in other cases the homicide may be unplanned but may occur because the weapon (e.g., kitchen knife, gun) is conveniently available. In still other premeditated situations, a third person is hired to do the killing. The presence or absence of children may be another significant variable, both in terms of their immediate presence at the time of the killing or in terms of the woman claiming that she killed the abuser for the children’s sake as well. A three-phase cycle of violence was proposed by Walker.4 It is based on a tension-reduction learning model that operates in intimate relationships. In this framework, violence between intimates is seen as always progressing toward a more serious level. Respites may occur from such events when one of the parties commences legal action or attempts


209

clinical interventions. Walker’s first phase is the building of tension, in which a woman attempts to exert some control over her life by trying to give the man what he wants in view of threats that he will beat her. Next, inevitably, is the period when the control no longer works and battering incidents begin to occur. The hypothesis is that the psychological release of tension in the relationship functions as a reinforcer of the cycle. The third phase is one of loving contrition, or simply no tension, which also serves as a reinforcer to the pattern and contributes to maintaining the relationship. Walker’s theory is based on the paradigm of a learned-helplessness model. The woman is exposed to repeated random and inescapable aversive stimuli. Eventually the woman loses confidence in her ability to predict whether she can control the violence by her own behavior. She adapts to the situation through cognitive distortions such as minimalization, denial, and splitting of the mind from the body during painful times. Such mechanisms have also been viewed as survival techniques.6 Those clinicians who use therapeutic attempts to interpret this behavior as random and to focus on the good side of the man are viewed as misinformed and using misguided therapeutic efforts.

Psychiatric Considerations The reasoning supporting the battered spouse syndrome is that a person who has been battered is experiencing PTSD or serious depression associated with the psychological impact of abuse. The procedure at trial is first to introduce expert testimony that the battered spouse syndrome exists in the case at hand. The case is then elaborated clinically by such expert testimony as a version of PTSD if a formal diagnosis is needed. Such an approach can raise questions about the validity and reliability of PTSD in general and more specifically about its use in battered spouse syndrome cases. These issues are then relevant to clinical discussions and to the context of the implications for these allegations in a courtroom. An important point to be underscored is that the PTSD diagnosis is not used for purposes of introducing an insanity defense or even for a diminished-responsibility defense. Rather, the diagnosis of PTSD or some other diagnosis for a battered woman is presented as an explanation and justification for the act in an exculpatory sense. The woman in question is usually portrayed as a woman who was terrified because of the past physical assaults perpetrated on her. Her emotional state is described as the normal response of a terrified human being to the abnormal and dangerous environment to which she has been exposed. There also seems to be a presumption that the most significant

210


diagnosis found in a battered woman in such a situation who commits a homicide is PTSD, or at least the only one seen as legally necessary. An opposing view would argue that the validity of the battered woman syndrome is based on inadequate research and theoretical inconsistencies. Because a key legal issue in trials using the battered spouse syndrome defense is the standard of self-defense, a crucial criterion is deciding what is meant by being “in imminent danger of serious bodily harm or death.” How continuous must a state of believing that one is in imminent danger be? If the atmosphere is more or less continuous, the advocates would argue that it does not matter whether the beating was recently or days ago. In response to the question of why the woman did not shoot just to wound or maim, the reply is given that such action would not resolve the woman’s ongoing fear of bodily harm or death once the man had recuperated. Therefore, it is argued that the woman had to kill the man to be relieved of the fear that he would kill her in the future. A common finding among the diverse contexts in which a battered spouse syndrome is used is that there was often either minimal or no psychological evaluation of the parties before the killing, testifying either to the illness or competency of the involved parties. Attempts at marriage counseling would not be included within the realm of psychological evaluation, because there is usually no preexisting assessment with respect to diagnosis or personality functioning. This lack of a competent psychological appraisal may actually be seen as a boon by the legal defense team because they would not want anything introduced that could pertain to the possibility of some abnormal mental state existing in the perpetrator that would detract from the issue of a justifiable homicide; the woman’s actions, seen as the actions of a victim having a battered spouse syndrome, are subsumed under a self-defense argument. If such an approach is used, clinical insight into the parties is bypassed entirely. Any past counseling that has been attempted is used not to focus on diagnostic insights into the parties but solely to focus on whether a battered spouse syndrome was present before the event took place and to update the validity or usefulness of this evaluation to the present case. If the homicidal act has already occurred, the consideration is whether the background provides sufficient evidence of victimization to argue for a defensible homicide. There are those who, in supporting a battered woman defense to homicide, see it as essential that a therapist become the victim’s advocate and wholeheartedly accept the victim’s position in contrast to adopting a neutral or value-free stance. This lack of neutrality is in opposition to traditional psychotherapeutic approaches and family therapy, and it blurs the distinctions between advocacy and clinical assessment.


211

Sociological Viewpoint The question of why a battered woman would kill her partner remains intriguing. The usual answer given is more of a sociological one, based on a position that men have been socialized into believing that they have a right to control women and therefore see it as permissible to use violence on women. The result is that there are battered women. This basic position may be generalized to seeing violence against women as the core of violence in the world. Such a gender-based psychology, or perhaps sociology, arose in the late 1960s in opposition to psychological theories of male and female behaviors as innate and biologically based;8 these theories were seen as biased both in their empiricist approaches and in how hypotheses were formulated and tested. Although this sociological focus was originally used in sex discrimination cases, it was eventually extended to oppression of women. Women who were subjects of violence were viewed as victims of coercion. A feminist perspective on problems of intergender violence provides a further insight into understanding this defense when such killings occur. In this perspective, wife abuse is generally seen as a reflection of power differences existing between the genders. Physical violence is viewed as the most efficient means of social control to maintain dominance over women.7 From this perspective, it would seem unnecessary to use other theoretical frameworks to explain extreme violence between partners. For example, a family systems approach, commonly used by those who work with families, would be interpreted as ignoring the crucial significance of gender as the key variable. For similar reasons, questions about the possible presence of psychiatric disorders in either party would be viewed as perhaps interesting but largely irrelevant. In the feminist perspective, comparisons have been made between the situation of battered women and that of a hostage who is kept under the control of a captor. Social institutions of marriage and the related patriarchal family structure are viewed as conducive to a form of captivity that permits and sanctions the use of force to control women. Rather than viewing the family as a safe haven from violence and intrusion from outside forces, the feminist perspective sees the family as a cloak for violence that can occur within it. Operating within this framework, traditional views are seen as inhibiting professionals from looking inside families to see how the system of dominance and abuse can operate. When a killing does occur within such a family milieu, it is used to confirm the view that a marriage always has the potential of ending in violence. As a corollary, advocacy is seen as necessary to reveal the state

212


of marriage as it actually exists. Advocacy is also seen as a necessary step to make such domination public knowledge. The situation of a woman standing trial for murder of her partner is viewed as a paradigmatic political situation through which an activist approach can reveal what actually transpired in a marriage under the veil of family secrecy.

Legal Considerations To transpose such a sociological view of the relationship between the genders into the legal setting once a homicide has occurred requires a change in how the standard of self-defense is interpreted. Whereas selfdefense has customarily been predicated on demonstrating facts that a person had a reasonable perception of some imminent danger, some jurisdictions have allowed redefinitions so that the standard shifts to how an “objective reasonable woman” or a “subjective reasonable battered woman’s perceptions” can be applied to these situations. The need is to portray the woman who kills as living in a state of perpetually fearing for her own life. Underlying this shift is the position that the danger of great bodily harm or death is always imminent for a battered woman. Another legal difficulty is the context in which the eventual homicide occurred. In one study of women who had killed a man, 86% were in an active, intimate relationship with the man at the time of the murder.9 In most typical cases, the battered spouse syndrome defense is raised where homicidal violence occurred in the midst of a struggle. However, this defense becomes more problematic when some time elapsed between a past violent episode and the homicide. Those people who advocate the syndrome as being exculpatory of homicide are not troubled by these difficult cases. They view the timing of when the act occurs as a relatively insignificant variable. The act is perceived as almost inevitable, barring some unforeseen event or major intervention measures by someone. Such a problem is handled by arguing that a reasonable woman perceives a need to defend herself with a weapon against a man who uses his body to dominate and injure her. Another difficulty emerges when a woman kills in a nonconfrontational setting. These are situations in which a mate was asleep, was lounging in a chair, had his back turned, or was in a state of stupefaction from drug or alcohol use at the time of the killing. In one case, in which a woman was raped by two acquaintances who then threatened to return and repeat the act, it was argued as reasonable for her to perceive she was still in a state of imminent danger when she shot one of the rapists hours later.10 A guilty verdict was then handed down because she was not considered to have been in imminent danger at the time. Sub-


213

sequently, an appellate court accepted the application of imminent danger to her situation in the sense that imminent did not mean immediate; at a retrial, the woman was acquitted on grounds of self-defense. The question of the admissibility of expert testimony supporting self-defense in the context of a battered spouse situation has been pursued vigorously since the Ibn-Tamas case in 1979.11 In that case, the court ruled that the knowledge provided by an expert on such situations was beyond the knowledge of the average juror, and it was more probative than prejudicial to allow such testimony. The expert witness was then seen as providing background data to help the jury make its crucial determination of whether the woman actually and reasonably believed her life was in danger when she killed her husband. Some appellate courts have held such testimony admissible, whereas others have rejected it.12 The U.S. Supreme Court has not ruled on the matter.

Implications of the Battered Spouse Syndrome Many implications arise from homicide trials in which experts testify that a woman who kills her partner under these circumstances was justified in her actions.

Implications on Society’s Image of Women How a female should be viewed and how well the criminal justice system functions are issues that have emerged. One view of women is that they are helpless creatures who cannot deal with force as it is used by physically larger and more muscular males. Although this argument is undoubtedly true for some women, just as when a smaller man has to confront a physically larger and more powerful male, an extension of the principle to other areas of the criminal law would imply a regression to a primitive societal state in which the weak and helpless either need a more physically powerful champion to protect them or routinely need to carry a weapon to equalize the situation so that they can survive. The image conveyed is that of the caveman giving protection to his woman at the price of the woman’s autonomy. In battered spouse syndrome defense cases, defense lawyers use people and experts to testify about a woman’s helpless state that was induced by her being subjected to continual battering. Although the old stereotype of the woman actually enjoying the abuse has largely been dispelled, the argument is that now another stereotype is being reinforced—that of the passive female victim in contrast to a person capable of taking charge of her own life and, if necessary, testifying to the state of the ongoing and actual troubled relationship with a partner.13

214


A general question, but one the jury must decide in trial, is which image of a battered woman is to be believed: that of a brainwashed, frightened victim or that of a woman who has done what any reasonable woman would do, in killing her oppressor after reaching a certain end-point. The former image is actually a clinical portrayal of a woman with many features of a personality disorder; the latter is a sociological image of a woman existing in an oppressive and male-dominated culture who decides it is time to stop the beatings to avoid living in fear. The inference is that because society has not created a different milieu in which she can live with a partner, she must take matters into her own hands. From this confused image, a jury may be led to a manslaughter or even an insanity verdict. However, advocates of the battered spouse syndrome defense want the verdict to be not guilty, based on the premise that the woman dealt rationally with a situation that required her to take self-defensive actions and not because she acted insanely. Furthermore, it is desirable to avoid the administrative complications (e.g., psychological treatment, medications, forced hospitalization) that follow a “not guilty by reason of insanity” verdict.

Validity of Battered Woman Syndrome Defense for All Battered Women There is also confusion when assessing the impact of battering on an individual woman who has subsequently killed the significant partner in her life. A persistent question is why all women in such predicaments do not kill. The battered spouse syndrome has been formulated as a theory about women in general who have been battered.14 However, the difficulty with it is that it does not explain the psychological idiosyncrasies of different personality styles and defenses among individual women. Matters are forced into an either-or categorization based on whether sufficient characteristics of a battered woman syndrome are present for any given woman. The result, as it is for allegations of PTSD in legal settings, is to make sure all the usual descriptive symptoms are listed. More troubling psychiatrically is the seeming compartmentalization of large numbers of women who have killed a spouse or lover into the same diagnostic format as though no differences exist in their personalities, motives, or diagnoses. The approach seems to take a constrictive view rather than encompass the diverse possibilities that operate with different diagnostic possibilities, such as when a male commits a homicide. It raises a question as to whether women are ever seen as killing a partner for some reason other than being victimized. Analogous questions arise when women who have been convicted of homicide and im-


215

prisoned are subsequently released by a governor’s pardon. A mixture of diverse motives for killing might well be present in such cases, varying from battered women responding in desperation to their plight to pecuniary motives or delusional possibilities.

SELF-DEFEATING PERSONALITIES AND KILLING: MYTH OR REALITY Personality disorders have never been free from some type of pejorative connotation; however, the debate regarding establishing a diagnosis of self-defeating personality disorder has been a particularly grueling and vicious one. Often the personality disorders are based on sociological theory. For example, some posit that certain groupings of behaviors and attitudes that they see as congruent with being primarily an adaptive response to a coercive environment are mislabeled as personality disorders. Regarding self-defeating personality disorder, supporters of the above view see these behaviors as adaptive to a milieu in which misogyny is seen as an endemic condition in a society and obedient behavior is a reflection of a traditional female role. Thus, these supporters hypothesize, the seemingly adaptive nature of a female who exhibits selfdefeating behavior is so prevalent that it is not recognized as a result of abuse. Specific criticisms of this diagnosis from a sociological perspective are as follows: 1. 2. 3. 4.

The diagnosis is gender biased and will be misapplied to women. A normative behavior in women will be subject to misinterpretation. The criteria reflect adaptation to victimization. There is an assumption that a person can choose to be removed from victimizing situations. 5. The diagnosis will be misused in courts.15 From a clinical standpoint, the diagnosis of self-defeating personality disorder has been criticized because it relies excessively on subjective criteria and lacks standardized instruments. Such criticisms could be raised about many diagnoses, especially in the personality disorder categories because of the lack of consensus on how to assess personality diagnoses in view of their correlation and continuity with normal functioning.16 The criteria for these diagnoses resonate with clinicians when a specific personality disorder with which they are familiar may have relevance in homicidal situations. Although the approach may have some validity, the question remains whether a diagnosis made on the clinician’s assessment of subjective components (e.g., assessment of the

216


patient’s internal motivational patterns) can meet the criteria for an acceptably valid diagnosis. When comorbidity is present with the diagnosis of self-defeating personality disorder, there can be considerable overlap with avoidant, dependent, and obsessive-compulsive categories. However, some clinicians believe the diagnosis is distinguishable from these other types of disorders.17 In essence, critics of the diagnosis of self-defeating personality disorder believe that the tools currently available for its assessment and diagnosis, as well as those of other personality disorders, do not equip clinicians to differentiate between someone who has a persistently disordered personality and someone who is reacting to an adverse social milieu. If that critique is accurate, the focus should logically be on the more pervasive presence of abuse as a societal and legal problem rather than on delineating specific traits within a group of people. Hence it would be concluded that clinicians who look for intrapsychic motivation in self-defeating behaviors confuse a result for a cause. However, this argument seems inconsistent in some respects. When a male assaults his female companion, his actions are not seen as solely the societal response of someone who is blindly caught up in exhibiting his male prowess. If a homicide results from such assaults, the psychiatric diagnosis of the perpetrator becomes of great interest. In fact, 80%–90% of those who assault their wives may have personality disorders.18 The concern of diverse groups was that self-defeating personality disorder could impinge on the validity of the battered spouse syndrome as a grouping under PTSD, given that masochistic personality was considered to be a legitimate diagnosis. In this argument, self-defeating personality disorder was attacked as a diagnosis that had no empirical basis and lacked any scientific validity.19 In contrast, these groups argued that sufficient scientific criteria existed for battered spouse syndrome to be included under PTSD.20 Lurking throughout the arguments was the thread that diagnostic formulations regarding masochistic personalities were more appropriate for those who had a primary and underlying disturbance that sometimes led them to kill their lovers rather than for those reacting under the primary influence of a PTSD brought about by being the victim of continual assaults.

Features of Self-Defeating Personalities The descriptive features that were proposed for a diagnosis of selfdefeating personality disorder pointed to a pervasive pattern of selfdefeating behaviors beginning in early adulthood and manifested in diverse contexts. The description “masochistic personality disorder” had


217

been abandoned to avoid historic implications from older psychoanalytic views that connected female sexuality with masochism or from the belief that women derived pleasure from being assaulted.21 Self-defeating individuals were seen as avoiding or undermining pleasurable experiences and being drawn to situations or relationships in which they would suffer, yet structuring matters in such a way as to prevent others from helping them. As included in the DSM-III-R22 appendix, eight diagnostic criteria were proposed for this disorder, five of which were needed to make the diagnosis (Table 7–1). If the diagnostic criteria are viewed behaviorally, they suggest an individual predisposed to becoming involved in situations in which the probability of self-destructive or violent behavior is increased. Those who have worked in the criminal justice system, or in forensic settings, will be immediately impressed with how many of these characteristics fit males in a court population, who make up at least 90% of criminal cases. Note the caveat in which the diagnosis is not to be used if the selfdefeating behaviors occur only in response to, or in anticipation of, being physically, sexually, or psychologically abused. Similarly, if the person is depressed and shows such behavior, the diagnosis does not apply. The key was to distinguish whether the self-defeating behavior was a persistent pattern as distinguished from situations of actual or anticipated abuse. For example, a person with many past destructive behaviors might only recently have become a physically abused victim. As might be expected, self-defeating behaviors are commonly seen in diverse personality disorders, such as borderline, dependent, obsessive-compulsive, and avoidant; these self-defeating behaviors also become entwined in some dysthymic and major depressive episodes. If the diagnosis of self-defeating personality disorder were to be used, it would have a good deal of comorbidity. Although not part of the criteria listed, exposure to physical, sexual, or psychological abuse as a child or being reared in a family in which spousal abuse occurred has been suggested to predispose individuals to self-defeating personality disorder. This proposal is similar to what has been argued by some researchers as the predisposing factor for development of borderline personality disorder. However, not everyone would agree that events in childhood are causally related to the emergence of a self-defeating personality disorder in adulthood. In Walker’s4 study of 400 cases of self-identified battered women, she did not find support for such a causative relationship; instead, she interpreted the relationship between these factors as simply one of coexistence. The behavior of the women was not seen as self-defeating in nature but rather as their effort to protect themselves as best they could. Walker felt that interpretations

218

TABLE 7–1.

HOMICIDE: A PSYCHIATRIC PERSPECTIVE DSM-III-R criteria for self-defeating personality disorder

A. A pervasive pattern of self-defeating behavior, beginning by early adulthood and present in a variety of contexts. The person may often avoid or undermine pleasurable experiences, be drawn to situations or relationships in which he or she will suffer, and prevent others from helping him or her, as indicated by at least five of the following: (1) chooses people and situations that lead to disappointment, failure, or mistreatment, even when better options are clearly available (2) rejects or renders ineffective the attempts of others to help him or her (3) following positive personal events (e.g., new achievement), responds with depression, guilt, or a behavior that produces pain (e.g., an accident) (4) incites angry or rejecting responses from others and then feels hurt, defeated, or humiliated (e.g., makes fun of spouse in public, provoking an angry retort, then feels devastated) (5) rejects opportunities for pleasure, or is reluctant to acknowledge enjoying himself or herself (despite having adequate social skills and the capacity for pleasure) (6) fails to accomplish tasks crucial to his or her personal objectives despite demonstrated ability to do so (e.g., helps fellow students write papers, but is unable to write his or her own) (7) is uninterested in or rejects people who consistently treat him or her well (e.g., is unattracted to caring sexual partners) (8) engages in excessive self-sacrifice that is unsolicited by the intended recipients of the sacrifice B. The behaviors in A do not occur exclusively in response to, or in anticipation of, being physically, sexually, or psychologically abused. C. The behaviors in A do not occur only when the person is depressed. Source. Reprinted from American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987, pp. 373–374. Used with permission. Copyright 1987 American Psychiatric Association.

of behavior patterns as self-defeating were based on two contradictory assumptions: that the behavior patterns 1) were part of an inherent selfdestructive pattern but 2) were identified as part of the criteria for diagnosing that self-destructive condition. It is paradoxical that because a woman is more likely to be the victim of a homicide at the time she attempts to terminate a self-defeating relationship, it would appear that her chances of surviving are better if she simply remains in such a relationship. Thus it could be argued that a battered woman who tried to adapt to a relationship was only trying to survive. Yet it could also be argued that her staying in the relationship


219

perpetuated her self-defeating pattern of behavior. In fact, the initial separation period is also likely to be the time when the potential for violence is maximized against not only the woman but also the man (perpetrated by the woman). A question is whether the price of freedom for an individual is that of increasing the risk of violence and possible homicide. By the time DSM-IV-TR23 arrived in 2000, the solution was to eliminate the possibility of a diagnosis, such as self-defeating personality disorder, or the consideration of such behavior as a valid diagnostic entity. The criteria set proposed in DSM-III-R was not provided for as an axis for further study (see Table 7–1). Furthermore, terms such as selfdefeating, sadism, and masochism are not listed in the “Glossary of Technical Terms.” Unfortunately, the omission or decision does not resolve the problem of people who may act in their own worst interest. Clinicians continue to witness such behavior in many of their patients. In some homicide cases, self-defeating acts may be one of the key variables that need an explanation to make sense of what happened.

Tentative Conclusions Regarding Self-Defeating Personality Disorder Some tentative conclusions can be offered from this continuing controversy about the diagnostic validity of self-defeating personality disorder in connection with homicide and the disorder’s official abandonment as a descriptive diagnosis. One approach would be to view the self-defeating behaviors as part of a comorbid picture of different personality disturbances, especially dependent or borderline personality disorder. In addition, self-defeating behaviors can be part of the clinical picture of depressions not evolving in response to an actual or anticipated physical abuse. Impulsive, angry, acting-out behaviors would then be viewed as part of a repertory of responses accompanying an existing depression. In a study using the Structured Clinical Interview for DSM-III-R and the Personality Disorder Examination, Skodol and colleagues24 found that self-defeating personality disorder overlaps substantially with the borderline and dependent personality disorders and raised questions about redundancy. The alternative would be that self-defeating behaviors are simply present in many diagnoses. A quite different approach is to interpret self-defeating types of behavior as part of PTSD, in which the response is expected given the distressful situation a person has been in. Yet many fundamental issues remain unresolved about PTSD because this category was created primarily for individuals who were reacting to major disasters, military combat, or criminal assaults.25 Epidemiological surveys continue to use these given diagnostic criteria and to find a high prevalence of this type of PTSD in communities.

220


It is not clear how PTSD relates as the immediate response to any given trauma and how the pervasive responses of anxiety and depression to stress should be distinguished from PTSD. Not only are there many subclinical variants of PTSD, but the course and duration of PTSD vary a good deal as well. A homicide victim or perpetrator of a homicide may often be assessed with this diagnosis.

UNDERSTANDING MASOCHISM AND HOMICIDE FROM A PSYCHODYNAMIC PERSPECTIVE Acknowledging the controversy surrounding even the validity of the diagnosis of a masochistic character does not therefore negate its heuristic value for raising explanatory hypotheses when a homicidal act has occurred. If the theoretical framework being used in explanations is made clear, and the inferential processes are specified, psychiatrists are not confined to viewing the immediate behavior as being simply reactive but can inquire into the intrapersonal conflicts as well as the interactional and social factors. Searching for an explanatory model in homicides requires one to focus on an individual who has already committed the act. Just as a detective does in murder mysteries, the psychiatrist must operate after the fact to reconstruct the events, but here the detective work is not to find the murderer but to explain the way the perpetrator’s mind operated the way it did. In formulating explanations of a perpetrator’s masochistic behavior, one should note that masochism is not used here in the restrictive sense of erotogenic or sexual masochism. These types of masochism involve various sexual acts in which pain is sought out as pleasurable. Although some of these sadomasochistic practices can lead to a death, it is usually by a mischance rather than a calculated act of homicide. These acts should be distinguished from the group in which killing accompanying some type of sexual predatory acts is the primary goal. The majority of masochistic acts that lead to an eventual homicide are more subtle. Their nature only becomes clearer when a detailed clinical assessment can be carried out and corroborated by outside information. This group belongs to the far larger grouping of masochistic behaviors in which characterological or “moral” masochism is operative.

Early Psychodynamic Theories Explanations of masochistic tendencies leading to violence have been offered in diverse theories and in a growing body of literature. Various schools of thought exist, and to complicate matters further, within each


221

school theories have shifted over time. Freud’s26 thinking on this topic was a classic example of changing theoretical viewpoints. Freud ultimately came to view masochism as the struggle between the ego, representing the life instinct, and another basic instinctual source—the famous death instinct; the instinctual nature of the struggle was often played out in the context of oedipal conflicts and defenses. Masochism was seen as deriving from the superego, with a defensive submission to the father as a defense against castration. That is what led Freud to label such traits as passive and feminine. Suffering and seeking punishment by overt or covert means became the masochist’s goal. True masochists were seen as those who never missed a chance to turn the other cheek when they had a chance to receive a blow. Many questions remain unresolved within one or another psychodynamic viewpoint. Many of the leading figures in psychoanalysis in the 1930s and 1940s wrote about these issues both within and without the classic Freudian framework.27 Controversy developed over whether and when pre-oedipal versus oedipal factors played a leading role in the formation of sadomasochism. More specific issues arose in the debate, such as whether a developmental arrest initiated the pattern of sadomasochistic personality or whether it resulted from a regression.28 Some viewed masochism as a continuum between normality and pathological expressions.29 The role of powerful parental figures split into good and bad images, with difficulty integrating them, entered the theoretical picture. When the self-concept involved a sense of “badness,” the devaluation was seen as related to the development of a masochistic submission on one hand or to the attribution of badness to others on the other hand. Since then, theories have gone in different directions. A child, and later an adult, experiencing pain or trauma and feeling helpless may develop defenses in an effort not only to try to control others but also to deal with the aggression that gets mobilized. Thus, theorists supposed that the experience of pain aroused aggression that would be directed toward those seen as responsible for the pain. In turn, defenses against the expression of aggression would develop. One outcome would be that aggressive behavior would be responded to in kind, setting the stage for repetitive aggressive interactions with others. Theorists conjectured that a subtle type of collusive pattern might evolve between assertion and submission, centering around these aggressive conflicts. Situations of child abuse illustrate this pattern in an overt way with helplessness and control issues. The development of the person’s capacity to fantasize is one possibility to limit the repetition and transform self-defeating patterns of aggression.30

222


Self-defeating behavior is then not perceived as simply the expression of an aggressive drive per se but rather is linked to impairments in the ability to overcome narcissistic injuries. People or things that the person sees as frustrating elicit destructive or murderous wishes. Defenses try to deal with the aggression that is mobilized in an effort not to lose key attachments. If violence emerges, it is viewed as the person’s response to severe and overwhelming threats to existing relational bonds and to narcissistic equilibrium.31 Succeeding generations of researchers have continued to grapple with these problems. In the following discussion, I do not recapitulate the fluctuating theoretical positions and shifts but rather take the approach of extracting something from the theories that is relevant to homicides. The question is how these individuals with a diathesis toward self-defeating interpersonal relationships, whatever the cause, might eventually end up behaving in a homicidal manner.

A Synthesized Psychodynamic Theory on Masochism and Homicide Many complexities within the individual ego and interpersonal life operate in masochistic individuals to predispose them to an ultimate homicidal act. These complexities involve a recurring set of behaviors in which some variety of self-defeating behavior, often of an unconscious nature, eventually sets the stage for a violent response. It is not the person’s eventual, overt act that offers the key to his or her motivation but the pattern of characterological traits that have led that person into humiliating or punitive relationships with other people. One common feature found in masochistic killers is that they have often submitted to positions or tasks that are emotionally painful in their daily lives; these might or might not be overt acts of abuse. Rather, the essential component is one of the person’s being trapped in an abject, submissive, or exploited position. The progression from being humiliated or punished to becoming the perpetrator of such actions is so subtle that only those who have known the person quite well throughout his or her lifetime can sense these patterns and their enduring quality. Three components are key to understanding the transition from being the masochistic object of aggression to being the perpetrator of such aggression—suppressed aggression, narcissistic tendencies, and depression. If one begins from the position that the vulnerability of persons with a masochistic personality structure to commit a homicide lies partly in the difficulties they have in dealing with their aggression, then several things begin to fall into place. Narcissistic aspects in masochistic phenomena play an integral part in the development of homicidal ten-


223

dencies. Kernberg32 argued that an understanding of the narcissistic aspects of masochism as related to these individuals’ hatreds and the psychopathology of aggression is just as important as focusing on sadistic tendencies and wishes. In the background lurks a proneness to depression. A confluence of these tendencies seems to operate in the cases that result in homicide. Some people become depressed when they realize their hatred is mounting and they cannot control others around them. They sense that any expression of their overt anger or even sadistic tendencies and wishes would elicit guilt over such acts of wrongdoing. One pattern found repeatedly in homicide cases has been that the person initially attempted to counter these tendencies or undo some aggressive acts. In intimate relationships, the perpetrator may attempt to undo the aggressive acts by engaging in exaggerated acts of niceness or being submissive. Problems emerge from a feeling of psychological coerciveness that is present in these self-defeating relationships. Masochistic individuals use diverse maneuvers to induce guilt in those on whom they are dependent, especially when they feel that their needs continuously remain unmet or they have provoked rejection. Whereas one outcome may be a depression in response to a sense of defeat, another may be mobilized anger. These individuals’ harsh self-assessment may be turned on those whom they consider responsible for their plight and the source of their unhappiness. Eventually, alternations between angry outbursts and unconsciously putting themselves into situations in which they are maltreated reflect the beginning of a breakdown in their ego controls, as well as attempts to deal with the unremitting and overcritical demands of their superego. Hence it is not inconsistent to see sadistic (“murderous”) attacks on people who have played key roles in their lives. These attacks reflect the individuals’ attempt to maintain their emotional well-being by shifting the demonstrations of humiliating behavior in an effort to regain control over their own lives. This shift signifies that on some level they are developing an awareness that techniques of control by suffering are not working. In the meantime, these individuals are eventually moving toward a state of desperation in which they conclude that it is time for them to do something about their predicament of being treated badly by someone else. These are not the same dynamics operating as when they seek external praise and admiration based on their narcissistic needs. This state of desperation rather reflects a thwarted sense of closeness and being taken care of that they sense is not going to be restored, despite the pain and suffering they have endured.

224


The ascension of these self-defeating and sadistic tendencies lowers these individuals’ threshold for a state of action, which leaves them at risk for a primitive and destructive striking out. Homicide then becomes a possibility, although other impulsive acts or a continual return to self-abasing behavior is more likely. Some people become assaultive, whereas others fall back on suicidal gestures or self-mutilations. Attempts to distinguish rage attacks occurring in these individuals from those seen in individuals with borderline personality disorder may become more of an academic exercise because the dynamic psychopathology overlaps. Pathetic attempts to regain control over others through self-abasement may be a last effort to thwart the full expression of their hatred. The experience of narcissistic mortification induces a frightening loss of control in these individuals. Both their inner and outer worlds seem beyond control. The threat of loss of control reveals such a degree of weakness in these individuals that it can be taken as a token of an imminent collapse or annihilation. The dual needs of regaining control over others and over oneself coexist. Denial or repression may transiently serve to contain the sense of being overwhelmed and helpless. Yet accompanying the state of paralysis from the mortification is that of feeling utterly humiliated because their grandiose self is overwhelmed. Eventually, seeking the source of such a blow, or projecting the terror and misery experienced onto an external object, allows their rage to become more focused on an “enemy.” When this object becomes focused on at the same time that these individuals sense an annihilation or even a psychotic decompensation, that is the crucial period for a homicidal act. Some type of victory is needed to triumph over an impending sense of devastation. Although suicide is one solution, a better solution may seem to be to destroy the source of their oppression. At this point, their superego functioning has moved from a position of restraint or guilty control to that of joining in a destructive act that has the quality of righteous assertion. Another variety of masochistic enslavement that can manifest itself in a homicide is seen with the problems connected to a delusional (paranoid) disorder, especially of the erotomanic type. Such cases (discussed in Chapter 3, “Schizophrenia, Delusional Disorders, and the Prediction Problem Regarding Homicide”) reflect acts based on delusions—for example, when the person holds the distorted belief that someone is keeping her beloved from her and then concludes that the interferer must be eliminated. The victim then need not be the person who is the object of the delusional attachment; the victim can be the person seen as doing the interfering. However, if these individuals delusionally come to be-


225

lieve that the person who supposedly loves them has altered his or her love for them, they may attribute the change to someone else perceived as malignantly changing the feelings of the beloved object. These individuals’ twisted altruistic reasoning then formulates that it is better for their former loved one to die than to be unable to reciprocate their love.

Stalking Patterns In some stalking cases, the masochistic individual has held the fantasized belief that some other person has been attracted to him or her for years, although that belief may have waxed and waned over the years; the masochistic person acts only after some mischance has pushed him or her into it. Celebrities or other well-known community figures are often the recipients of such special attention. They may have been stalked over time, but this may be revealed only after some threatening or homicidal act has occurred. Unsigned letters may be sent to those being stalked in which hostility is interwoven with allusions to the mutual destiny of the letter writer and the addressee. Letters may refer to the writer’s suffering and torment, with accusations of how his or her heart is being torn apart or life is being ruined unless some acknowledgment is given that his or her feelings are being reciprocated. The John Hinckley case, involving the actress Jodie Foster, would appear to fit into this category. Many of these people’s lives have been uneventful, and they have often lived in relative isolation, not becoming conspicuous until some public act brings them into the public eye. Sometimes their persistence and forced attentions cause complexities and embarrassments for their families and places of employment, but these developments are fortuitously better than an outcome of homicide. A more subtle, and consequently more difficult, situation to detect is related to self-defeating patterns in a love relationship that is not reciprocal or in which one partner wishes to withdraw. Here we are not usually dealing with delusions per se, but rather the dynamics of sadomasochistic features in people who continue to be attracted to those who flatly reject them. When these situations emerge abruptly, rather than within the context of long-standing and conflicted relationships, the possibility of a manic episode intervenes. For example, the individual seems to abandon his or her past commitments and norms and neglect many other aspects of his or her ongoing life to pursue someone who may be only mildly interested in or even indifferent to the individual. Narcissistic aspects are operative in the individual’s self-absorption with this one person. To suffer the pains and disappointments of such a pursuit seems to enhance the pride of the sufferer.

226


The more common situation is where a love relationship once existed but the feelings of one of the parties have changed, as in the following case example.

Case Example A 30-year-old man lived with his female companion for 6 years. Together they had a 5-year-old child; the woman’s older child from a previous relationship also lived with them. After 6 years, the woman informed her partner that she no longer wanted to continue the relationship. His first reaction of denial slowly gave way to one of incredulity that her feelings could have changed. However, acquaintances felt that the woman had become bored with him over the preceding 2 years. The man refused to leave the apartment, and 1 year later his companion obtained a court order to put him out of the apartment. Despite this prohibition, he periodically reappeared, which eventually led to contempt of court charges. When a new man moved in with the woman, the tempo of the original boyfriend’s behavior increased, and the potential for violence became magnified. One night, after having stalked the woman at her apartment for some time, the man burst into the apartment during a party and shot his former partner, her new male friend, and both children.

What explanation can be offered for such killings? Often they are simply referred to as revenge killings, yet this explanation seems insufficient. One reason for these types of killings is that the person often has not resolved his or her continued idealization of the previous love object. In addition, to a point, such individuals endure the suffering of being excluded and rejected. In the case just described, the man suffered for an entire year for the sake of the woman who had rejected and become indifferent to him. In some cases, the rejecting person is more sadistic in his or her rejecting behavior. During the period of rejection, the man in the case example would talk about his anguish to anyone who would listen, including various therapists. At times, his depression and suicidal ideation were discussed. His continued idealization of the woman contained elements of distortion in seeing her as being more financially successful than she actually was. He believed that through her he would share in her idealized attributes and attainments, and his beliefs interfered with his ability to loosen his attachment to her. When he saw that his grandiosity was not going to continue to be realized through her, and no one else was to be considered as having value for him, he resorted to a final destructive act. Suicide subsequent to homicide is always possible in these cases. This man did not commit suicide but simply went to a nearby restaurant he often frequented and drank coffee


227

while awaiting his arrest. However, it could be said that serving the three consecutive terms for murder to which he was sentenced was a suicide equivalent. In dynamic terms, this man’s grandiose self would not let him literally destroy himself.

Malignant Masochism In some cases, committing a homicide might be just a more final and climactic failure reflecting diverse areas of difficulties in the person’s life. A final self-destructive act could unconsciously reflect the ultimate in self-denial. These individuals might be bound up in conflicts for years over controlling others and being controlled. The external trappings give them the appearance of being controlled and manipulated by a spouse, lover, employer, or other person. Yet as noted earlier, victimization is one more method of control by way of a distorted display of aggression. A perplexing dynamic in such submissiveness is the individuals’ need to induce guilt in those who maltreat them. In the long run, these attempts are usually unsuccessful. The mechanism appears related to the self-righteousness present in masochistic individuals’ control system. The mission is to change the other party—much like a person who wishes to point out the mistakes of those once presumed supportive of him or her. Asch33 referred to the extreme end of such self-destructiveness as malignant masochism because of the intensity of the self-destructive components with prominent hostility. In these people, even more than in other self-defeating individuals, gaining control over others becomes the primary mission in life. The battle commences at a young age and may acquire sophistication as part of a set of character defenses throughout development to adulthood. The pattern of achieving success by being a failure or thwarting others is commonly seen. Sometimes the battle over control is fought within, as witnessed in periods of drug or alcohol abuse. These individuals often stress their ability to control their usage of substances simply through willpower. Similar variations can be played out through bulimic or anorexic behavior patterns. Externalization of the conflict onto their environments is done through skillfully provoking or manipulating others into seemingly maltreating them or inciting others to carry out such acts. In one case, subsequent to the homicide, the perpetrator’s employment record revealed that this individual had instigated investigations of a series of grievance complaints at several jobs against several people. Matters had dragged on for several years without resolution by committees that, in retrospect, appeared simply to dismiss the complaints. However, the very nature of these entanglements became a

228


source of pride for this person because they allowed a demonstration of suffering to become evident from frustrated attempts to make others accept guilt over the person’s various predicaments. Even in legal cases in which a trial or settlement finds in these individuals’ favor or in which some grievance judgment is made against another party, the individuals experience only short-term relief. This response may occur because the other party is not perceived as experiencing sufficient suffering, but only anger and resentment. The complainant continues to feel that something unjust has been done to him or her that needs to be righted. The pleasure in controlling, at whatever price, leads to repetitive role enactments with similar themes in diverse settings. Those clinicians who attempt to treat these individuals, by whatever therapeutic modality, also witness the similarity in repeated therapeutic failures. Lurking in these reenactments is always the threat of retaliation against these practitioners who have not seemed sufficiently empathic with the suffering experienced by the patient—and who thus need to be punished. When depressive components are also present, the masochistic individual courts a combination of punishment from both within and without. In the context of these backgrounds, a homicide may occur in response to an internal conflict, with superego functioning insufficiently ameliorative, and based on failed needs to externalize and punish others. On one hand, the person needs to appease a punitive and overcontrolling source from within that ultimately knows no surcease. At the same time, he or she needs to find some culprit in the external world and attack and punish that person. The final scenario is a striking out, in what is often portrayed as a variation of self-defense, provocation in a fight, or an attempt to overcome a thwarted grievance. There is often a diagnostic overlap by descriptive signs and symptoms and by psychodynamic formulations. In other cases, a feeling of failure emerges if a key person, seen as the culprit who is somehow responsible for the person’s unhappiness, has gained distance from the entanglement. This removed distance must then be punished. In the following case example, the balance in the relationship tipped so that the person who was initially enslaved by her insecurities became less conflicted and wanted to extricate herself. Such a transition left the partner feeling confused and betrayed. In this stalemate, a rage reaction led to a homicide.

Case Example A young couple in their 20s had lived together for 2 years. At the time they met, she was insecure and wanted to be rescued from her parental home. In the beginning of the relationship, he saw himself as the victim of her unfounded jealousies and haranguing about women with whom he


229

worked. Although these accusations never totally ended, a shift occurred when she no longer needed him to meet many of her needs. His discomfort mounted when he felt he had progressively less to offer her and sensed her reduced need for him. Because his role of maintaining her selfesteem had become less important in the relationship, he resorted to other forms of self-sacrificing, such as letting her routinely drive their car to work while he rode a bus. Anger at being her “puppet” emerged. His attempts to induce guilt in her for his sufferings—attempts that had formerly led to reconciliatory moves—became less effective. When she announced she was going to move back with her parents, he became physically violent for the first time. Feeling progressively helpless to control her, he initiated arguments and would not relent. Public scenes occurred, with one or both of them walking away from the other. After an argument, he would drink to excess, something he had not previously done. His justification was that the arguments provoked him to drink. He was experiencing two areas of failure: not being in control of the current situation and a failure of what he had expected to be a long-term relationship that would lead to marriage. The police had been called to their home on two occasions for domestic disputes when neighbors heard her screaming during assaultive episodes. On a third occasion the police did not arrive in time; the woman had already died from intracranial hemorrhaging. The man had now achieved the ultimate in punishment and in subjugating himself. It is problematic whether a freedom from internalized controls that induced him to act out his murderous rage succeeded in his need to continue to think that she was the one who needed rescuing. His unresolved needs to obey internal edicts to provoke and subjugate others remain untouched and continue to be played out in a location par excellence for reenacting such dramas—a state prison. Only at the price of relinquishing his sense of self could he maintain the belief that he had no role in his own fate.

HEALTH CARE PROFESSIONALS AND HOMICIDE Clarification of Issues Those who work in the health care area are in a special position to commit homicidal acts should they deviate in that direction. When this occurs, it is accompanied by widespread publicity and notoriety. It is thus important to keep in mind some basic distinctions. First, the distinction must be made whether some type of professional relationship has been established, such as a doctor–patient relationship, and whether the person was in a health care treatment context when the act was carried out. Deaths that occur through medical negligence are not part of a criminal homicide. Second, if the homicide is perpetrated by a health care professional, but it is not carried out in the course of any professional work, it may fit in with other types of homicide discussed in chapters of this book.

230


Physician-assisted dying and voluntary euthanasia are also areas that need to be distinguished. These deaths touch an ongoing debate in the political, moral, and theological arenas that continues not only in the United States but internationally. Health care personnel are apt to have as diverse opinions on these issues as the general population. In the background of any discussion on physician participation in dying is the co-opting of German physicians during the Nazi regime, where a memorandum from Hitler allowed physicians to kill disabled people. About 200,000 people were subsequently killed in psychiatric asylums.34 Depending on the circumstances and the country, physician-assisted dying may or may not be legal and hence may or may not be viewed as a homicide. In recent years the discussion has been stimulated by physicians assisting in suicide, such as Dr. Jack Kevorkian, who assisted in 100 cases of suicide in Michigan until that state passed a statute prohibiting assisted suicides. He was convicted under that statute on a ruling that there was no constitutional right to suicide, and the verdict was upheld by the Michigan Supreme Court, which affirmed suicide as a common law crime.35 Kevorkian later participated in the television program 60 Minutes via a videotape in which he gave a lethal injection; he was then sentenced on a second-degree murder charge. These situations usually arise with individuals who are suffering from terminal diseases, such as some form of cancer, amyotrophic lateral sclerosis, leukemia, or lymphoma. Another situation with notoriety was that of Dr. Timothy Quill. He wrote in a leading medical journal about his decision to provide pills sufficient for suicide to a female with terminal leukemia.36 Similar issues and publicity attend cases in which individuals are accused of assisting in a suicide by helping a loved one who is terminally ill to die. Two cases that raised constitutional issues reached the U.S. Supreme Court regarding assisted suicide. In Vacco v. Quill, the Court held that a state statute criminalizing assisted suicide did not violate the equal protection clause.37 In Washington v. Glucksberg,38 the Court held that state prohibition of assisted suicide did not violate the due process clause. These rulings have left matters with terminal sedation as the alternative to physician-assisted suicide, accompanied by a withdrawal of hydration and nutrition. In this confused state of affairs, the movement to enact dignified death provisions arose with Oregon passing the first such proposed law in 1994, allowing physician-assisted suicide under specified conditions.39 On one hand, this law has led to efforts to legalize assisted suicide in other jurisdictions, but on the other, it has also led to countermeasures from those who are concerned about technology taking over at the end of a person’s life. The issue is posed


231

in terms of allowing a terminal patient to have a death with dignity by way of advanced directives and the right of a competent person to refuse unwanted medical treatment as a protected liberty interest.40 This situation led U.S. Attorney General John Ashcroft, in 2002, to direct agents of the Drug Enforcement Administration to take punitive action against physicians who prescribed medication at the request of a terminally ill patient who wished to hasten his or her death.41 However, on an academic level, the debate has continued around issues such as conflating termination of artificial life-support systems with the right to have another person assist one in suiciding.42 Some have also argued that there is really no need for physician-assisted dying because for all but a few cases, palliative care is available. In response, others argued that it is for these few cases that the issue becomes important.43

Health Care Killers The material just discussed has been presented to set off types of homicidal acts carried out by health care professionals in medical settings under professional auspices. Although homicide charges may arise in some of the situations described, these acts are carried out either under political sway or in the context of testing the limits under which a physician acts in the belief that he or she is entitled legally to help patients. The discussion that follows has some overlap with these situations but extends quite widely beyond any clinical or legal justification. When these cases are studied, diverse motivations for homicidal acts are seen. Some of the cases involve confused ideas of assisting others; some have been based on misreading the patients’ or their loved ones’ wishes in the context of suffering; and others have been based on the narcissistic wishes of the physician-perpetrator or may simply be homicides carried out by health care personnel. Some of the latter cases might be considered serial killings done under medical auspices. Certainly not all of these cases involve sadomasochistic personality disturbances, but enough of them do to merit being discussed in this chapter. However, other primary disorders may be present as well. At worst, some of the episodes illustrate a patient’s misplaced trust and a physician’s disregard for others. In some cases, a monetary motive has a role, but other cases would qualify as serial killings. Despite increasing publicity in many countries about homicides by staff on hospital wards and in nursing homes, there is a lack of scientific studies on the subject. We are basically still at the level of reporting challenging cases. Great ambivalence is evident in descriptions of these events. Thus on one hand, the offenders are labeled as monsters, killers,

232


witches, and so on, but on the other hand, they are then sometimes referred to as saviors who save terminally ill patients from prolonged suffering. One study focused on cases involving 20 medical homicide offenders in various parts of the world. These offenders comprised 14 males and 6 females; 16 offenders were nurses, of which 10 of the cases were based on scientific publications or newspaper articles and 6 on legal documents.44 Cases involving family ties were excluded as well as those in which greed was the primary motive. None of the offenders was acquitted by reason of insanity or found to have a serious mental disorder. The number of victims varied between 2 and 50; many more homicides could not be proven. After the first homicide, inhibitions were lowered to perpetrate again. One of the primary personality characteristics found in the offenders was a prominent self-consciousness. The assessment was that they appeared unable to distinguish reliably their own perceptions, moods, and feelings from those of the patients. As a result, the offenders’ own states of personal suffering and apathy were transferred to the patient. This transference was correlated with an inability to assimilate sorrowful experiences. The offenders seemed to depreciate suffering and dying people, as well as the offenders’ own personality, anticipating the same or worse conditions for their own prospective suffering and dying. The perpetrators described pity for the victims and confused that with self-pity when devotion to a patient led to no relief for the patient or themselves. The offenders’ idea was that they then would perform an act of mercy, supposedly for the patient. The author raised a question as to whether this identification with the patient is connected to the depersonalized milieu of large numbers of people dying in hospitals and nursing homes, often attended by inexperienced staff, and to the perpetrators not wishing to die themselves in such circumstances. In the milieu of suffering, where hopelessness and dying become a daily routine, the suggested vulnerability for some persons is that death ceases to mean anything special. It is not as though those in the helping professions throughout history have not on occasion engaged in homicidal behavior based on different motives. Iserson45 elaborated on several key cases occurring in different countries. His data indicated that 45% of the killers in the health care professions were nurses and 25% physicians, with the remainder distributed among dentists, opticians, nurses’ aides, nursing home workers, orderlies, and midwives. The majority of cases occurred in hospitals, nursing homes, and other residential facilities. The most common means of carrying out these types of killings in the past were by poison,


233

but at the present time a variety of drugs are used, such as potassium chloride, pancuronium bromide, digoxin, and succinylcholine.46 The ease with which physicians and health care workers have access to drugs to carry out killings differentiates them from others who commit homicide. Iserson45 classified physicians who were murderers into three groups: 1. Serial killers. Those who carry out serial killings in the context of medical settings based on a misplaced trust of the victims. The case of Dr. Michael Swango is often mentioned as an example.47 He pled guilty to murdering three patients in the Veterans Hospital in New York and a young woman when he was an intern at Ohio State University Hospital and is currently serving a life sentence. Dr. Swango kept a diary in which he said he killed for pleasure and enjoyed the smell of an “indoor homicide.” This group of medical killers raises questions about why a person becomes a serial killer. 2. Political killers. Physicians who participate in corrupt political regimes in which they directly or indirectly participate in homicides. This classification would hold for any physician participating in acts of political killing. Some physicians in Japan and Nazi Germany during World War II fulfilled this description. They participated in experiments on prisoners that would lead to the prisoners’ deaths or on prisoners who were deemed the “undesirables.”50 3. Episodic killers. Episodic killings are perhaps related to a perpetrator’s mental disorder or are committed for highly personal reasons. A multitude of diagnoses and motives seem to apply in these killings—involving love, hate, greed, psychosis, psychopathy, fear, and anger. Most victims are the perpetrators’ family members, friends, rivals, or coworkers. These perpetrators typically kill only once or twice. Some notorious cases belong to this group, in which physicians commit a homicide based on their disturbed emotional lives, just as other distraught persons who may commit homicide.

Case Examples The case of Harold Frederick Shipman, a British physician, illustrates one of the more egregious cases of medical homicide.48 It is believed that he actually killed many more of his patients than the 15 women for whose deaths he was criminally prosecuted and found guilty in 2000. Many of these, but not all, were elderly women, and their deaths seemed connected to his injecting them with heroin. A significant background factor might have been his taking care of his sick mother during his adolescence, when she suffered greatly from lung cancer, relieved

234


only by her physician’s daily visits to administer increasing doses of morphine. A police psychiatrist suggested Shipman was a necrophile obsessed with the act of inducing death and thus having it under his control. Throughout many subsequent investigations, Shipman refused to participate or admit guilt. A public inquiry chaired by a senior highcourt judge, at a cost of about $40 million, found that Shipman had murdered 215 patients and possibly as many as 260 over 23 years. Shipman suicided in prison in 2004.48 An American case involved a nurse, Charles Cullen, who told prosecutors in 2004 that he had killed perhaps 40 people by intentional drug overdoses.49 These occurred over a 10-year period in hospitals in New Jersey and Pennsylvania. On one occasion, two people were killed on the same day. The drugs he used for killing were digoxin, insulin, nitroprusside, norepinephrine, dobutamine, and pancuronium, a medication that paralyzes critically ill patients so they can breathe through a ventilator. Some of the victims were terminally ill, but others were recovering. In agreement for a plea of 13 life sentences to spare him the death penalty, he agreed to cooperate with prosecutors in five other counties where he had worked as a nurse, to help them determine whom he had victimized there. Cullen’s explanation was simply that the killings were mercy killings, yet several victims were young or recovering. In his background were a history of being fired several times from hospitals and a number of hospitalizations himself for mental illness. Over the years, Cullen had been accused by patients’ family members of killing their loved ones. The pattern of killing was that the defendant would look up records of a patient who was assigned to another nurse and then withdraw unauthorized medications from a prescription cart that had a built-in computer to monitor who was dispensing which drugs. In the counties that had no computer systems for medications, reconstruction had to rely on the information provided by the defendant. Before his nursing career, Cullen had been in the U.S. Navy and served aboard a nuclear submarine, at times spending more than 2 months submerged in the Atlantic Ocean. An attempted suicide led to his discharge from the navy, and he later obtained nursing training. Over 11 years he had had nine jobs, usually being forced to leave for vague reasons that were not disclosed to subsequent employers. The murder of Dr. George Parkman by his Harvard Medical School colleague, Dr. John White Webster, in 1849 is a notorious American classic. Webster had borrowed money from his colleague and classmate, Parkman. An argument ensued when Parkman discovered that Webster had pledged a collection of gems—already pledged to Parkman—as collateral for another loan from a different party. Parkman was hit on the head and cut up as a corpse in Webster’s laboratory, adjacent to the anatomy laboratory. Webster was subsequently found guilty and executed.51 A more recent case is that of Dr. John Kappler, an anesthesiologist with a history of multiple psychiatric problems, initially of depression. Later,


235

he developed paranoid ideation that the Central Intelligence Agency was stalking him, that his house was bugged, and that his wife was trying to poison him.52 The records indicate that in 1980 he stopped a patient’s heart by injecting excess lidocaine, and that in 1985 he unplugged a quadriplegic patient’s ventilator, but the “attempted murder” charge was dropped from a lack of evidence. The culmination of Kappler’s acts came in 1990 when he drove onto a jogging path, where he hit and then carried the victim several hundred feet on the hood of his car; the victim happened to be a psychiatrist, not initially identified. After hitting this victim, Kappler then drove into a woman returning from grocery shopping and similarly carried her for several hundred feet on the hood of his car, causing her broken bones. In response to auditory hallucinations, he then wandered about, took a bus to New York, and tried to strangle his son. He was convicted of second-degree murder and is serving a life sentence in Massachusetts.

REFERENCES 1. Coen SJ: The Misuse of Persons: Analyzing Pathological Dependency. Hillsdale, NJ, Analytic Press, 1992 2. Dunne D: Menendez justice. Vanity Fair 57:108–118, 159–166, 1994 3. Rusbult CE, Martz JM: Remaining in an abusive relationship: an investment model analysis of nonvoluntary dependence. Pers Soc Psychol Bull 32:558–571, 1995 4. Walker LEA: The Battered Woman Syndrome. New York, Springer, 1984 5. Browne A: When Battered Women Kill. New York, Free Press, 1987 6. Walker LEA: Psychology and violence against women. Am Psychol 44:695– 702, 1989 7. Yllo K, Borgad M (eds): Feminist Perspectives on Wife Abuse. Beverly Hills, CA, Sage, 1988 8. Maccoby EE, Jacklin CN: The Psychology of Differences. Palo Alto, CA, Stanford University Press, 1974 9. Mann CR: Getting even: women who kill in domestic encounters. Justice Quarterly 5:33–51, 1988 10. Bochnak E (ed): Women’s Self-Defense Cases. Charlottesville, VA, Michie Company Law Publishers, 1981 11. Ibn-Tamas v United States, 407 A2d 626 (DC 1979) 12. Gillespie C: Justifiable Homicide. Columbus, Ohio State University Press, 1989 13. Schneider EM: Describing and changing: women’s self-defense work and the problem of expert testimony on battering. Women’s Rights Law Report 9:195–222, 1986 14. Crocker PL: The meaning of equality for battered women who kill men in self-defense. Harv Women’s Law J 8:121–151, 1985

236


15. Firester SJ: Self-defeating personality disorder: a review of data and recommendations for DSM-IV. J Personal Disord 5:194–209, 1991 16. Oldham JM: Personality disorders: recent history and future directions, in The American Psychiatric Publishing Textbook of Personality Disorders. Edited by Oldham JM, Skodol AE, Bender DS. Washington, DC, American Psychiatric Publishing, 2005, pp 3–16 17. Nurnberg HG, Siegel O, Prince R, et al: Axis II comorbidity of self-defeating personality disorder. J Personal Disord 7:10–21, 1993 18. Hart SD, Dutton DG, Newlove T: The prevalence of personality disorder among wife assaulters. J Personal Disord 7:329–341, 1993 19. Walker LEA: Terrifying Love: Why Battered Women Kill and How Society Responds. New York, Harper & Row, 1989 20. Comments: a critique and proposed solution to the adverse examination and problem raised by battered woman syndrome testimony in State v Hennum. Minn Law Rev 74:1023–1061, 1990 21. Widiger TA: The self-defeating personality disorder. J Personal Disord 1:157–159, 1987 22. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987 23. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 24. Skodol AE, Oldham JM, Gallaher PE, et al: Validity of self-defeating personality disorder. Am J Psychiatry 151:560–567, 1994 25. Davidson JRT, Fou B: Posttraumatic Stress Disorder: DSM-IV and Beyond. Washington, DC, American Psychiatric Press, 1992 26. Freud S: The economic problem of masochism (1924), in The Standard Edition of the Complete Psychological Works of Sigmund Freud, Vol 19. Translated and edited by Strachey J. London, England, Hogarth Press, 1961, pp 155–170 27. Grossman WI: Notes on masochism: a discussion of the history and development of a psychoanalytic concept. Psychoanal Q 55:379–413, 1986 28. Glenn J, Bernstein I: Sadomasochism, in Psychoanalysis: The Major Concepts. Edited by Moore BE, Fine BD. New Haven, CT, Yale University Press, 1995, pp 252–265 29. Brenner C: The masochistic character. J Am Psychoanal Assoc 7:197–226, 1959 30. Grossman WI: Pain, aggression, fantasy, and concepts of sadomasochism. Psychoanal Q 60:22–52, 1991 31. Rizzuto A-M, Meissner WW, Buie DH: The Dynamics of Human Aggression. New York, Brunner-Routledge, 2004 32. Kernberg OF: The psychopathology of hatred, in Rage, Power and Aggression. Edited by Glick RA, Roose SP. New Haven, CT, Yale University Press, 1993, pp 61–79


237

33. Asch SS: The analytic concepts of masochism, in Masochism: Current Psychoanalytic Perspectives. Edited by Glick RA, Meyers DI. Hillsdale, NJ, Analytic Press, 1988, pp 93–115 34. Burleigh M: Death and Deliverance: ‘Euthanasia’ in Germany, c. 1900–1945. New York, Cambridge University Press, 1995 35. People v Kevorkian, 527 NE2d 714 (1994) 36. Quill TE: Death and dignity: a case of individualized decision making. N Engl J Med 324:691–694, 1991 37. Vacco v Quill, 521 US 793 (1997) 38. Washington v Glucksberg, 521 US 702 (1997) 39. Satz AB: The case against assisted suicide reexamined. Mich Law Rev 100:1380–1407, 2002 40. Cruzan v Director, MO Dept of Health, 497 US 261 (1990) 41. Lowenstein E, Wanzer SH: The U.S. attorney general’s intrusion into medical practice. N Engl J Med 346:447–448, 2002 42. Kamisar Y: The rise and fall of the “right” to assisted suicide, in The Case Against Assisted Suicide: For the Right to End-of-Life Care. Edited by Foley K, Hendin H. Baltimore, MD, Johns Hopkins University Press, 2002, pp 69–93 43. Nuland SB: How We Die: Reflections on Life’s Final Chapter. New York, Knopf, 1994 44. Beine KH: Homicides of patients in hospitals and nursing homes: a comparative analysis of case series. Int J Law Psychiatry 26:373–386, 2003 45. Iserson KV: Demon Doctors: Physicians as Serial Killers. Tucson, AZ, Galen Press, 2002 46. Hickey EW: Serial Murderers and Their Victims. Washington, DC, Wadsworth, 1997 47. Stewart JB: Blind Eye: The Terrifying Story of a Doctor Who Got Away With Murder. New York, Touchstone Books/Simon & Schuster, 1999 48. Esmail A: Physician as serial killer: the Shipman case. N Engl J Med 352:1843–1844, 2005 49. Kocieniewski D: Former nurse pleads guilty in death of hospital patients. The New York Times, April 30, 2004, p A21 50. Lifton RJ: The Nazi Doctors. New York, Basic Books, 1986 51. Schama S: Dead Certainties. New York, Knopf, 1991 52. Ablow KR: The Strange Case of Dr. Kappler: The Doctor Who Became a Killer. New York, Free Press, 1994


CHAPTER

8 THE DEPRESSED OR BIPOLAR PERSON AND HOMICIDE

UNRAVELING THE WEB of events and motivations that lead someone with a depressive or bipolar disorder to commit a homicide presents a major challenge. People’s perception of the link between depression and the acts leading to a death poses an interesting paradox: whereas suicide is automatically assumed to be related to a depressive disorder, the same relationship is not assumed when homicide and depressive disorders are considered, even in the view of psychiatrists. Typically, people assume no persuasive argument for such a relationship. Similarly, the idea that someone in the grips of a manic state might actually kill someone may seem strange to some clinicians. These responses occur particularly among those who have not assessed the depressed or manic people who have actually engaged in serious acts of violence other than attempted suicide. In this chapter, I appraise the problem of homicide and depression or bipolar conditions and offer a specific hypothesis about homicides in which there is an accompanying psychotic state. A two-pronged approach is necessary to understanding the relationship that may exist between depressions and homicidal behavior: one prong relies on epidemiological data and the other on clinical data based on direct contact with the perpetrators of homicides. Both approaches contribute to psychiatrists’ knowledge about homicides, although different knowledge bases are used with their respective limitations. Outstanding texts dealing with depression have not contained any discussion about the problem of homicide.1–3 Similar omissions of discussions about

239

240


depression and violence occur in review articles, although a few refer to violent episodes in manic patients.4 Rosenbaum and Bennett5 examined 17 randomly selected psychiatric textbooks and found that 11 did not mention homicide at all, and only 2 mentioned homicide as having a possible connection to depression. Yet the problem of suicide is routinely discussed in psychiatric textbooks, and although the number of homicides in the United States has fluctuated since 1980 from 16,000 to 29,000, homicide connected to depression or bipolar states is rarely discussed in texts. Furthermore, the risk of death by homicide in the United States is about two-thirds that of death by suicide.6

LINKING DEPRESSION AND HOMICIDE: THE NEED FOR DIAGNOSTIC SPECIFICITY The possibility of a relationship between some type of depression and homicidal violence is simply a variant of the same question that can be raised for different diagnoses. Findings must initially be predicated on the validity and reliability of a diagnostic nomenclature as well as the sensitivity and specificity of the diagnostic system and the diagnostic instruments used. One vexing problem in accurately diagnosing depression in homicidal individuals is that often a person is given a depressive diagnosis only after he or she has committed a homicidal act. The diversity of types of depressions raises questions as to whether such homicides should be viewed as having any connection with the depressed state at all, or whether more diagnostic specificity should be required.

Different Types of Depressive Diagnoses People who engage in homicidal behaviors often carry different affective diagnoses. Some have the diagnostic characteristics of dysthymia, whereas others appear to be experiencing a major depressive episode, and still others exhibit homicidal violence in the course of an acute manic episode. Even though these different hypotheses can be proposed to explain the relationship between homicide and each type of mood disorder, it is essential that a sufficient explanatory framework be developed to guide clinical approaches, whether those approaches are based on a descriptive nosological approach or a psychodynamic framework. Even if individuals with a mood disorder are assumed to be a homogeneous group, the possibility of comorbidity among those individuals always exists. There is a high rate of overlap among individuals between a depressive diagnosis and other comorbid disorders, particularly sub-

The Depressed or Bipolar Person and Homicide

241

stance abuse or dependence and some types of personality disorders. For example, a large number of those with borderline personality disorders often act impulsively and also carry depressive diagnoses. Such comorbidity not only leaves academic questions unanswered but also promotes confusion in trying to resolve legal issues. A question arises as to whether the majority of depressed persons who commit a homicidal act carry a predisposing personality diathesis, and whether a “purer” sample of depressed individuals would not be as statistically likely to commit a homicide but instead might commit suicide. These queries mean that when epidemiological data are relied on, the element of diagnostic uncertainty must be kept in mind because the diagnoses may be taken as a given without reference to the degrees of uncertainty and comorbidity that exist. Individuals with a lifetime diagnosis of recurrent depression also experience a substantial number of manic or hypomanic symptoms.7 The question then is whether the unipolar condition or the bipolar disorder is primarily related to the homicide, assuming one of the disorders bears some relationship to the act of killing.

Neurotic Depression Versus Endogenous Depression Epidemiological perspectives suggest that a certain type of depressed individual may be most prone to engage in homicidal behavior and be in a high-risk group. This suggestion raises questions about predisposing factors for a person’s vulnerability to committing homicide. In the quest for specificity, clinical indicators suggest that a person with increased risk would be in the midst of a psychotic depression or be caught in a depression that is progressing in severity into such a state. However, scholarly discussions pay minimal attention to this subtype of mood disorder and its relationship to violence. The topic is usually discussed in the context of endogenous depression, a term that emphasizes vegetative signs and symptoms such as psychomotor retardation, sleep disturbances, weight loss, difficulty in concentrating, and anhedonia and perhaps alludes to a different quality of mood disturbance. Originally, a distinction was made between endogenous depressions and neurotic or reactive depression, a distinction based on the severity of the depression.8 However, a major environmental event was found in the background of those with severe depressions just as readily as for those with less severe states of depression. Furthermore, the term neurotic, originally identified with reactive types of depression, was seen as passé and no longer a valid appellation. If precipitating factors are prominent, homicides may operate equally in both of these depressions—whether cast as endogenous or reactive.

242


The debate on the validity of the endogenous/neurotic distinction has shifted to indices involving prognosis and outcome, response to treatment, and biochemical studies. Complicating matters are studies whose results contradict what were once thought to be well-validated outcome results, in which endogenous depressions were supposedly associated with a poorer prognosis and higher morbidity than neurotic depressions. In their study, Keller and colleagues9 found a high degree of chronicity across all types of depression despite modern treatments, with 12% of the sample patient population they studied not recovering during a 5-year period. To complicate matters, neuropsychological testing of patients with psychotic depression showed a pattern of dysfunction that was similar to that of patients with schizophrenia.10 Also, a subtype of major depression includes individuals who experience “anger attacks” (30%– 40%) and have a predilection for hostility, anxiety, and somatization. These individuals’ actions are seen as inappropriate in situations compared with a group that does not have anger attacks.11 Furthermore, electroencephalographic and functional neuroimaging studies using induced anger have shown the ventral prefrontal cortex either involved in controlling impulsive outbursts or not mobilizing the prefrontal cortex when needed for control. Those individuals who are predisposed to anger and aggressive outbursts exhibit decreased activity in the prefrontal cortex on neuroimaging studies compared with control subjects.12 A relevant question then becomes what the increased potential is for homicide in the group with unremitting depression and one of these diatheses.

Significance of Delusions and Hallucinations In a group of depressed individuals with homicidal potential, the presence of delusions and hallucinations is significant in addition to the standard signs and symptoms of depression listed in DSM-IV-TR. 13 Some researchers simply refer to this type of depression as a delusional major depression, because it appears that the hallucinations are less frequent than the delusions. This subtype of depression reflects a recent widening of the scope of the disorder from earlier classifications; in the past, individuals with these symptoms were often given a schizophrenic or schizoaffective diagnosis. The classification of patients with mood-incongruent psychotic affective illnesses is important.14 These patients would appear to possess the characteristics of a group of depressed individuals with a higher homicidal potential. A paradigm shift has occurred on how to view psychotic symptoms in depressed people. By the time of DSM-III,15 the process had been reversed from diagnosing affective illness by way of excluding schizo-


243

phrenia to diagnosing schizophrenia after the exclusion of an affective illness. The result was a category of mood-incongruent psychotic disorder. A labored attempt was made to give schizoaffective disorder its own criteria consisting of conditions that did not meet the criteria either for schizophrenia or a mood disorder. At one time, patients with this diagnosis were viewed as presenting with both a schizophrenic and a mood disorder, whereas at another time the criteria consisted of psychotic symptoms without mood symptoms. Schizoaffective disorder was accurately described in DSM-III-R16 as one of the most confusing and controversial concepts in psychiatric nosology. In DSM-IV-TR, schizoaffective disorder is described as an uninterrupted period of illness during which a major depressive, manic, or mixed episode is present concurrently with symptoms that meet criterion A for schizophrenia. A great variety of temporal combinations is then possible in terms of the prominence of different symptoms. When the question is the diathesis toward homicidal behavior, the concept of psychotic major depression seems more parsimonious. It is not unusual to think of psychotic behavior when a bipolar disorder with a manic episode is diagnosed. Results from one study indicated that manic patients may be more likely to show psychotic symptoms than patients with depression.17 In another series of 68 manic patients, it was found that 47% had delusions, hallucinations, or both.18 These delusions are usually grandiose, but they may also have a paranoid component. Although the paranoid delusions may be an early symptom of mania, it is often difficult to assess the delusional quality of assertions when they are made in an expanded mood. Hallucinations also often have a grandiose theme, and then the diagnosis can be particularly difficult to delineate from paranoid conditions. In this context, voices may be suggesting or commanding violent acts. If the homicide victim is an intimate, there is a tendency to attribute the motivation for the homicide to jealousy. Two approaches predominate in descriptive attempts to diagnose psychotic depressions. One stresses the presence of delusions and hallucinations; the second assesses the severity of the depression dimensionally by rating symptom severity. However, the key seems to be the pattern of hallucinations and delusions specifically related to guilt or hypochondriacal concerns, which clinicians tend to underestimate as patterns of psychotic thinking. Estimations of the frequency of psychotic depression vary because of the lack of strict criteria as well as its failure to be recognized as a distinct syndrome. The qualifier “with psychotic features” is added in DSM-IVTR, indicating a thought disorder with delusions or hallucinations. Thus

244


psychotic depression is not conceptualized as a separate entity, although Schatzberg and Rothschild19 noted the uniqueness of psychotic depression versus nonpsychotic depression based on computerized literature searches in which the biology, family transmission, course, outcome, and treatment in the two groups were compared. The distinction is important, because evidence indicates violence is partly associated with delusions in the mentally ill.20 In practice, different definitions may be used to diagnose psychotic depressions, along with varying levels of scrutiny of psychiatric symptoms. As a result, estimates of the prevalence of psychotic depression vary from 16% to 54%.21 If the hypothesis being put forward has validity—that homicides perpetrated by depressed people primarily come from the psychotic subgroup—then classification and proper use of the concept are crucial to understanding and predicting homicidal phenomena. This subgroup also possibly has a poorer treatment response to standard antidepressant treatment, which increases the likelihood of homicidal behavior over time. A variety of symptoms may be present in a psychotic depression that may in the future have predictive implications for homicidal behavior. Some delusions in psychotically depressed individuals have a paranoid quality, with ideas of reference, suspiciousness, and persecutory themes. Others have a guilt-ridden quality of being sinful or a nihilistic component. Somatic delusions are a third variety of delusions whose significance is often missed, because people who have them may be considered as being merely hypochondriacal. More than half of psychotically depressed people experience more than one kind of delusion.22 Auditory and visual hallucinations occur with equal frequency, and even tactile and olfactory hallucinations, customarily thought to be more associated with organic pathology, can be present.23 Other symptoms in psychotically depressed individuals are severe psychomotor agitation or depression, prolongation of the depression, ruminating qualities, and a failure to respond to treatment, all of which carry the risk of an unpredictable and explosive outcome. Also, dissociative states can possibly occur in individuals with or without a history of early abuse or trauma. Psychotic symptoms may be experienced during these states, and a homicide can occur during depersonalized states, with later partial amnesia. Assessment is complicated in such cases and, if organicity is ruled out, the question of malingering frequently arises. Another perspective is to view depressions along an extension from nonmelancholic depressions to melancholic and then to psychotic depressions.24 The melancholic group has one clearly specifiable difference from nonmelancholic depression—the presence of psychomotor disturbances. These disturbances may progress in psychotic depres-


245

sions not only to delusions and hallucinations but also to the person becoming mute or denying psychotic features altogether. Outbursts of homicidal acts that surprise those who know a person may come from this group of silent, delusional brooders.

EPIDEMIOLOGICAL QUESTIONS AND DIRECTIONS FOR RESEARCH Until recently, there was a seeming dichotomy in perceptions of mentally ill individuals in comparison with the general population: the general public and politicians perceived mentally disturbed people as more violent, whereas researchers in the social and behavioral sciences did not detect significant differences between mentally disturbed people and control groups. Advocacy groups for those with mental disorders supported the latter findings. Such advocacy occurred outside the realm of scientific discourse. However, the resultant confusion has had an impact on social policies, in legislation, and in courtrooms. The epidemiological approach does not focus on the specifics of how any particular depressed individual committed a homicide, nor does it seek to elicit specifics about the final common pathway eventuating in homicide. The pathogenesis of such behavior is bypassed not because it is insignificant but rather because the epidemiological line of inquiry is interested in answering a different type of question. However, the presupposition of an epidemiological approach is that the diagnoses leading to a homicide do have validity; lacking validity, the conclusions of that approach become suspect. In this respect, the prevalence of depressive episodes with depressive symptoms in the general population is relatively frequent and affects 4 of every 1,000 individuals.25 Feelings of worthlessness or guilt were seen as indications of the presence of psychotic features. For those experiencing manic episodes, about half have psychotic features.26 These psychotic groups compose a high-risk category for those who may commit acts of violence. Given a valid diagnosis, the search for a significant relationship between homicide and psychotic depression can go in one of two directions. In the first approach, a group of depressed or bipolar people are selected who have been given some type of specific depressive diagnoses; they are then tracked for any subsequent incidence of homicidallevel behavior. The homicidal-level term means behavior that could easily have resulted in a homicide, even if a legal homicide did not actually result. Hence many acts that would be legally classified as aggravated assaults would be included, such as shooting someone but not killing him or her because a vital organ was just barely missed. Limiting the

246


group to only those who successfully completed a homicide would result in a literal and tighter subject group. Although such a definition would lend more specificity to the search, its constrictiveness would bypass many perpetrators’ behaviors among depressed individuals that are behavioral equivalents to completing a homicide. Another troubling problem with this approach is how to define a perpetrator as being depressed. Even assuming diagnostic validity, should the depressed group be confined to those with a major depression, or should the group also include those with a bipolar disorder? Should the group be broadened further to include those with a dysthymic disorder as well? What about those in a state of abnormal mourning who engage in homicidal-level behavior? One solution would be to classify each subgroup under the general heading of being depressed until a large enough sample was obtained. The traditional approach— of relying on hospitalized patients rather than those attending outpatient clinics or of attempting to use a random community sample— poses methodological problems. Studies of homicidal-level behavior in this model are likely to focus on hospital populations and only occasionally include some clinic groups. In the second approach, the search begins with a group that has already been classified as homicidally violent because populations are chosen from jails or correctional facilities. The search would then retroactively examine the incidence of depression or manic states with psychotic features among those individuals in an attempt to assess their clinical state at the time of the homicide (similar to determining a person’s mental state at the time an offense was committed). One practical problem encountered in formulating this framework is the difficulty in obtaining an adequate sample of such individuals who have engaged in these extremes of violent behavior. Only a small number of those accused or even convicted of a homicide are psychiatrically examined in the United States, in contrast to many European countries, in which the majority of such a group is examined.27 Monahan28 pointed out three types of studies in which the relationship between having some type of mental disorder and violent behavior has been assessed in hospitalized mental patients. From these three types of studies, an analogy can be made to the specific case of some type of depressive disorder and homicidal violence. One type of study focused on the retrospective analysis of data for acts that occurred before the person was hospitalized. A second type tabulated violent behaviors that occurred while the person was in a hospital setting. A third type used follow-up studies that tracked the incidence of violence after a person was released from a hospital. When like kinds of studies are


247

grouped together, the following statistics are found: results from 11 prehospitalization studies indicated that a prevalence of 10%–40% (median = 15%) of subjects committed acts of violence shortly before hospitalization; results from 12 hospital studies also indicated that a prevalence of 10%–40% (median = 25%) of subjects committed acts of violence while on the hospital ward; and the results of the most recent posthospitalization studies (although there are studies dating back to the 1920s) indicated that 25%–30% of male patients had a violent episode within 1 year after discharge. 29 However, all of these studies lacked diagnostic specificity. Using an empirical approach, the first step is to find those who have already committed a violent act, and the second step is to determine the incidence of some preexisting type of mental disorder. If mental disorders do contribute to violent behavior, then the supposition is that the prevalence of mental disorders should be higher in this population than in a control group. One traditional method used with this approach is to study the incidence of mental disorders in incarcerated individuals; studies using this approach have found that the rate of incidence varied between 5% and 16% depending on whether a referred or random sample was used. (Of course, there are violent people who have not been institutionalized, but when it comes to homicidal violence, only the few who have not been apprehended are at large in communities.) Teplin30 used data from the California Department of Corrections to compare estimates of the prevalence of mental disorders among a group of prison inmates in California, a group of Chicago inmates, and the general population (using data from the National Institute of Mental Health’s Epidemiologic Catchment Area [ECA] study22). One finding from that study was that schizophrenia was 3 times more prevalent in the prison and jail populations, and major depression was 3–4 times more prevalent than in the general population. The prevalence of bipolar disorder was 7–14 times higher in prison and jail populations than in the general population. A confounding variable in such an approach is that mentally disturbed individuals are more easily apprehended or give themselves up more readily than “normal” criminals. Yet they also may be more easily diverted to hospitals, even in cases of homicide, and particularly if they have a history of mental disorder. Another limitation that also needs to be kept in mind is whether the diagnosis was actually manifest at the time of the homicide, in contrast to the diagnosis occurring earlier in the life history of a person. The advantage of mandatory psychiatric examinations for all individuals accused of a homicide was seen in some Scandinavian studies. One dealt with 64 individuals in northern Sweden who committed a homicide

248


between 1970 and 1981.31 Thirty-four (53%) were diagnosed as having a major mental disorder, and 38% of that subgroup also had substance abuse problems. An interesting additional factor is that 85% of the 64 perpetrators were intoxicated at the time of the homicide. In a Copenhagen study conducted over 25 years that focused on just homicide offenders, it was found that 20% of the men and 44% of the women were diagnosed as being psychotic.32 The presence of psychosis increased the risk of homicide for the men by a factor of 6 and for the women by a factor of 16. In this psychotic group, 41% of the men and 13% of the women had substance abuse problems, and 89% of the psychotic men and 21% of the psychotic women were intoxicated at the time of the homicide.

Limitations of the Studies Even considering the general question of the relationship between mental disorders and violence, these studies do not provide any answers with respect to the specific diagnosis of depression or bipolar disorders. Almost all lacked comparative data for the rates of violence among a nonhospitalized control group. The prime defect in all of the studies is that they focused on patients—either in the period shortly before they became patients or after their release. At whatever time period these mentally disordered individuals were studied, they all carried the hallmark of a group selected on some basis for hospitalization. Almost no research has focused on the possible connection between violence and mentally ill individuals who are outpatients in either public or private settings. In addition, most studies come from public hospital settings and none from private hospital wards. For the follow-up of released hospitalized patients, there is an additional bias because to be released from public institutions, these patients had to be judged as not being dangerous—otherwise, they would not have been released. Such a basis for release is especially germane, given the current legal situation of civil commitment being predicated on a standard of dangerousness as well as the possible legal consequences for mental health professionals who release patients who later commit a violent act. Finally, for violence that occurs during or after a hospitalization, it is not possible to detect what relationship may exist between the violence and a mental disturbance present before any hospitalization occurred. It is thus possible that some variables predisposing mentally ill individuals to violence may have been operating independently of the mental disturbance. These strictures hold for mental disorders in general. There is still a need to gather specific data on the prevalence of violence among individ-


249

uals who are not in treatment or have not been institutionalized but who may be depressed. One study attempted to reach these nontreated populations by using a database pool of 10,000 subjects from adult households in Baltimore, MD; Durham, NC; and Los Angeles, CA, that were part of the ECA study.33 The study authors used the Diagnostic Interview Schedule (DIS)34 to assess these individuals. Included were five questions dealing with violence; four of these were actually criteria applicable to the diagnosis of antisocial personality disorder, and the fifth related to the diagnoses of alcohol abuse and alcohol dependence. The authors described this test as a “blunt” instrument in the sense that the answers were based on self-reports by the subjects, and there was some overlap in the questions themselves. For those whose assessment allowed some type of DSM-III diagnosis, the rate of violence was five times higher than for those who did not merit a mental disorder diagnosis. Even more specific for the depression and homicide question was the finding that the prevalence rates for violence among those who received a diagnosis of a major mental disorder (i.e., schizophrenia, major depression, bipolar disorder) were similar. Most striking was the finding that the rate of violence for individuals meeting diagnostic criteria on the DIS for alcohol abuse or dependence and drug abuse or dependence was 12 and 16 times greater, respectively, than the rate of violence for those who had no diagnosis of a mental disturbance. These violent types were mainly young males from the lower class with a substance abuse problem and a diagnosis of a major mental disorder. The lack of diagnostic specificity plagues epidemiological approaches. The few sound studies that exist do not go beyond breakdowns into major mental disorders and substance use diagnoses. In another study, the authors used the DIS to assess 495 Canadian male penitentiary inmates and found that 35% of those who had at least one homicide conviction had been diagnosed with a major mental disorder in their lifetime, compared with only 21% of those who had no homicide conviction. Again, the comorbidity problem blurs these findings, because 83% of the mentally disordered group also had a history of alcohol abuse or dependence, and 64% had drug abuse or dependence. Congruent with the ECA data were findings from a study that used the Psychiatric Epidemiology Interview to find a relationship between symptoms and the occurrence of certain life events.35 In that study, a group of 400 people from a particular area of New York City who had never been in a mental hospital or sought help from a mental health professional were compared with a sample of mental hospital patients from that same area.36 Many variables were controlled for, including the homicide rate of the census tract where the subjects lived. The initial finding

250


was that the patient group was two to three times more violent than the never-treated sample. However, the researchers then did a further analysis in terms of current symptoms. When the factor of current psychotic symptoms was controlled for, no difference emerged between the two groups. In other words, the difference in rates of violence appeared to be attributable to the presence of active psychotic symptoms. If the patient group was not experiencing active psychotic symptoms, its violence rate was not much different than the control group from its home community; conversely, when either group was manifesting psychotic symptoms, the risk of violence increased. The inference for the depression-violence question is that in assessing depressed individuals for a tendency toward violence, whether they are experiencing a psychotic major depression is an important factor to be considered. It would also imply the importance of recognizing psychotic depression as a distinct syndrome, as Schatzberg and Rothschild19 did on the basis of statistically significant differences between psychotic and nonpsychotic major depressions.

Significance of the Studies At the minimum, the data suggest the significance of a psychotic major depression in the tendency toward violence. It is more likely that such individuals will behave violently than someone who lacks such a diagnosis. If a comorbid condition of drug or alcohol abuse is also present, the risk for violence becomes magnified. Certain types of personality disturbances, such as borderline personality disorder, which have not been studied in the context of epidemiological studies on violence, may also magnify the risk of violence, just as alcohol or drugs do. It is important to note that data from epidemiological approaches do not imply that because someone once had a psychotic episode, he or she is at a perpetually increased risk to commit a homicidal act. Instead, the increased risk of violence is linked with recurring psychotic episodes. It is significant that the overwhelming number of people with mental disturbances are not violent, and more specifically, even those who are psychotically depressed or in a psychotic manic state do not usually commit a violent act, let alone a rare event such as a homicide. At best, this statistically significant finding is interesting and leads researchers in a different direction than many earlier impressions that suggested no significant difference between mentally disturbed individuals and a control population. Most violent and homicidal behavior would still be related to other significant variables (e.g., age, gender, socioeconomic level). However, being in a psychotic state should presumably be considered a significant variable as well.


251

THE MURDER-SUICIDE PHENOMENON Researchers who have tried to distinguish a psychotically depressed from a neurotically depressed group on the basis of suicide attempts or suicidal ideation have obtained diverse findings. In cases of murder followed by suicide involving couples, Rosenbaum37 found that 75% of the perpetrators were depressed compared with a group that had committed a murder without suicide. Yet the assessments of depression were based on the perpetrators’ self-reports and the observations of a journalist; furthermore, Rosenbaum did not attempt to assess the type of depression. Although the murder-suicide combination is a relatively rare phenomenon in the United States, with a rate of only 6.22 occurrences per 100,000 people, it does shed some light on the vulnerability of psychotically depressed individuals.38 Coid39 found the percentage of homicides involving a suicide in the United States to be only 4%, whereas in Denmark it was 42%. However, the rate of murder-suicide tends to remain similar, perhaps because of a similar prevalence of mental illness in different countries. The principle seems to be that the higher the rate of homicide in a country, the lower the proportion of murder-suicides. The percentage of homicides followed by suicides is high in Denmark because of its low homicide rate. With a low homicide rate, a higher percentage of this smaller number is mentally ill. In a larger population, such as in the United States, there are fewer mentally ill in the group and hence a lower percentage. However, in Northern Ireland, with its continued civil strife, the inverse relationship was found between homicide and suicide.40 In West’s41 1960s study in England and Wales, 1 in 3 murders was followed by the suicide of the murderer, and only 1 in 100 suicides was coupled with a previous murder. This data is reminiscent of old observations about the relationship between aggression and depression, in which depression was seen as aggression turned inward.42 A more common type of murder-suicide is seen in a couple where one member is dying. The other member decides that he or she will end the suffering of his or her partner, and he or she carries out a homicide followed by his or her own suicide. The perpetrator does not care to survive alone and often leaves a message to that effect.

PSYCHODYNAMIC HYPOTHESES ABOUT HOMICIDES IN AFFECTIVE DISORDERS How the manifestations of a major depression or manic state with psychotic symptoms affect a person’s thinking has great relevance in deter-

252


mining the relationship between the clinical state and violence. A psychotic disturbance in thinking is not the only psychopathological hallmark of a severe affective disorder, but it is one significant variable. It is also highly relevant to legal questions that may arise in connection with psychotic affective disorders. Many outcomes other than homicidal behavior are possible for those psychotically disturbed individuals, depending on many chance variables (e.g., the type of treatment they receive). However, for these perpetrators and their victims, some hypotheses about how their disturbed thinking can lead to homicidal actions can be offered. The hypothesis is that when a depressed person commits a homicide, the act is likely to be connected specifically with the presence of psychotic thinking; more specifically, the homicide is most likely a response fostered by a thinking disturbance. One way this response is demonstrated is in the dissolution of how affects are processed by the person. In time, a secondary problem emerges in the modulation of the person’s actions. Before the psychotically depressed person commits a homicide, his or her thinking is dominated by feelings of dejection, guilt, and worthlessness. It is easy for even experienced clinicians to miss the transition into a homicidal state. The difficulty in spotting this transition appears to be related to assessing when a person moves from a state of sadness, which can be empathized with as a response to disappointment, into a state of delusionality. Another confounding factor in identifying a shift into a homicidal state is the psychotically depressed individual’s tendency to brood over some supposed wrong, which could be confused with a paranoid condition. Yet no overall fragmentation is present to confirm the paranoid state, and the thought disorder is focused on his or her own shortcomings more than those of others. When the psychotically depressed person progressively withdraws into a state of brooding about his or her shortcomings, a crucial transition has taken place. Too dramatic an interruption of this preoccupied state can mobilize homicidal behaviors. For those in a psychotic manic state, their thinking is also contaminated by grandiosity and a sense of righteousness that impinge on their decision-making capacity.

Case Example A 40-year-old man had stopped excessive consumption of alcohol 5 years previously when treated for a major depressive episode. Subsequently, he took pride in his changed life pattern and set high goals for himself. Three years later, economic conditions led to the failure of his company. When his efforts to find a new job led to only part-time employment, he became reimmersed in his old pattern of becoming ab-


253

sorbed in his failings. As often happens in cases of suicide when they are viewed retrospectively, his statements and references to his wanting to die and blaming others for his misery were not taken seriously. One day, during an argument with his daughter, he struck the girl and, becoming angry at himself, put his fist through a window, thereafter leaving blood splattered around the house. When his wife returned home, she called the police but then left the house out of fear for her safety. Meanwhile, the man had contemplated committing suicide by putting a gun to his head several times, but each time he was unable to pull the trigger. His broodings were abruptly interrupted by bright lights shining through the windows and loudspeakers through which the police were telling him to come out. He was instantly galvanized into action with energy he had not experienced in months and began a shootout with the police. The exchange of gunfire went on for several hours, resulting in one deputy being killed. Eventually, a sharpshooter’s bullet wounded the man so that he could not shoot. During his psychiatric assessment, the man indicated that he had intended to keep shooting to the point where officers would break into the house, gunfire would be exchanged, and he would be killed. Instead, he was charged with murder.

When a psychotically depressed person’s anger is mobilized, it allows him or her to externalize the self-hatred that normally immobilizes him or her. It is not uncommon for a severely depressed person to engage in behavior that is challenging or provocative to the police or those in some governmental agency (e.g., pointing a weapon at them, taking a hostage) in order to elicit a specific response. In some cases, the perpetrator may kill his or her entire family in this manner and then attempt suicide. This phenomenon is known as familicide; it is not uncommon for perpetrators in a large number of these cases to have a psychotic depression before the killings.43 One of the striking features seen in retrospect when a depressed or bipolar individual has engaged in homicidal behavior is that those close to that person sensed a prolonged mood of dejection or irritability in him or her that their attempts at reassurance failed to alter. Such efforts may have in fact elicited anger, which can be understood as the person resisting efforts to shift his or her delusional beliefs. In contrast, individuals with nonpsychotic depressive disorders may actually seek out reassurance. When deep-seated guilt in the psychotically depressed person progresses to a delusional level, a violent or suffering fate is expected if not courted. Given a delusional basis of assessing himself or herself as worthless, whatever adverse outcome occurs seems just. To take one’s own life is also fitting, but some cannot do that for diverse reasons, which then leads to violence on a broader scale. These individ-

254


uals may expect such a cataclysmic fate or even structure its happening. Patients with agitated features seem most prone to anticipate a disaster. If by chance they are seen by a mental health professional at that time, they articulate their sense of looming disaster in terms of a deserved punishment. If the punishment is not forthcoming, they do not experience relief. The psychotic depressions manifested by signs and symptoms of psychomotor retardation can go to the extreme level of the person being almost immobile, with accompanying vegetative signs; however, when the person’s anger becomes more available for action, it is likely that he or she will respond not only with suicidal but also with homicidal behavior. At this point, the person’s withdrawal into despair has lifted enough for him or her to destroy others or himself or herself.

Case Example A professional woman in her 40s was employed as a successful business executive. Her success and driven qualities were simply viewed by family and friends as part of her personality. A shift occurred when she first began to have religious experiences that puzzled her, but she later came to believe they were connected with her being selected by God for some special mission. Initially, she would see tears flowing from the figure of Christ on the cross, but these experiences expanded to hearing God’s voice reassuring her. She began buying expensive gifts for friends and believed she had the power to determine which people were connected to the devil and his works and who were righteous. When she got to the point of spending several nights awake and seeing angels outside her window communicating to her, the family took her to a clinic for an emergency appointment. While there, she believed the counselor seeing her at the clinic, along with certain family members, was going to stop her from carrying out her mission by putting her in a hospital. She then jumped into a car and drove away at high speed. The police were notified and a high-speed chase ensued. Her goal was to get to a church, where she believed she could claim sanctuary and would not be prevented from carrying out her mission. During the chase, police cars attempted to force her off the road. The result was that both the woman and the police officer lost control of their cars, and the woman’s car hit and killed a pedestrian walking along the street. The woman was charged with a homicide.

Emergence of Delusional Thinking I have referred to delusions as one of the central features in a psychotic depression, along with certain types of hallucinations, which can include command hallucinations for destructive acts. Delusions are infrequently elaborated in clinical reports on depression, which may be one reason why the diagnosis of psychotic depression involving homicidal cases is


255

underreported. It has been noted that some clinicians are reluctant to recognize psychotic forms of depression.44 Goodwin and Jamison1 noted that although only about 20% of psychotically depressed patients report having had hallucinations, this rate increases to 50% when the individuals become manic. Whether they take action on these beliefs depends on multiple variables, such as the severity, fluctuating nature, and specific content of the hallucinations. Five dimensional aspects of delusions may also contribute to whether action is taken: conviction, extension, bizarreness, disorganization, and pressure.45 When there is a preoccupation with an intrusive belief, the possibility that the person will act to relieve the situation is increased.46 The delusional system of psychotically depressed individuals may reflect their poor self-esteem. However, problems with self-esteem are present in diverse disorders. The difference with those in a psychotic depression appears to lie in the prominence of the degree of hopelessness. If they believe themselves to be beyond the pale of ever being redeemable and accepted as worthwhile, it is not inconsistent for them to engage in murderous behavior; in fact, such action is actually seen as congruent with their internalized image of badness. Paranoid twists can be added when they come to believe that others see them as they do. Perhaps they feel that all concerned need to die. The theme of punishing others as equivalent to punishing themselves emerges as a measure of meting out justice for all. In the 1960s, a series of articles were written that described the “sudden murder” as a single, isolated, and unexpected episode of violent behavior.47–49 The murders were seen as serving no purpose in the sense of personal gain or profit. Thirteen male murderers so classified were compared with 13 habitual criminals and 13 sex offenders; almost every diagnosis but psychotic depression was considered for the murderers, despite the fact that they described experiencing an incubation period characterized by feeling increased pressures to conform, an increased need to blame others, and progressive isolation. The relationship of these men with their mothers was described as ambivalent, with an underlying hostility; they also reported having hostile, distant fathers. Strong dependency needs with suppressed anger dominated their development. In that era of DSM-I,50 psychotic depression was rarely diagnosed, with the preconception toward schizophrenia or passiveaggressive personality diagnoses, although in retrospect many of those so diagnosed appear to have been severely depressed. Brooding fantasies that others have betrayed or abandoned them become mirrored in the enormity of these individuals’ delusional beliefs, such as their seeming responsible for the misery of others or the

256


miserable state of the world. To have been the cause of such ongoing misery in others, without a seeming explanation except their own badness, puts them in a mental state comparable with that of a murderer. If they kill someone else or commit suicide, they may experience a sense of relief. The exaggerated degree of hostile thinking in a deluded person carries the potential for action because the unresolved internalized anger and hate leave the person in a precarious state. Whereas paranoid individuals are more likely to turn on the environment initially, psychotic depressive individuals struggle to the point at which they are convinced that destroying those around them is an option. This conclusion often lends an air of calculation to their actions. Their resulting action may be an attack on intimates, a major act of violence at their place of employment, or a dramatic outburst in a public setting that puzzles those who lack knowledge about the perpetrator’s inner anguish. Turning hate against others in a regressed state reveals the perpetrators’ degree of ambivalence and hate. The hate may be turned on those whom they blame for their failure to achieve or to reach some pinnacle of happiness or success. The breaking point may come when they experience an increasing amount of unrelieved guilt in connection with the hate they feel toward loved ones or those on whom they are dependent. In some cases, when a person is betrayed in a love relationship, the third party is murdered, which allows the perpetrator’s denial system to maintain that it was not his or her loved one who was responsible for the pain and suffering he or she had been experiencing.

Catathymic States These explosive outbursts of homicidal violence suggest the idea of catathymia that Wertham 51 publicized, but which was actually introduced earlier by the German psychiatrist Maier, and perhaps even earlier by Kraepelin.52,53 Wertham described the state in which a person begins to believe that a violent act is the only solution to relieving his or her state of chronic emotional tension. An antecedent of prolonged conflict finally gives way to an act of violence, following which a sense of relief is experienced. Wertham54 later viewed catathymic crises as connected with delusional thinking and the presence of a psychotic state. More recent work has discussed a chronic form of catathymic violence in which an incubation period of depression is the most common affect before a violent act.55 The incubation period may last for days or months, with an accompanying disturbance in formal thinking as part of an obsessional brooding about murder or suicide. In addition to the


257

chronic form, the authors noted an acute catathymic state in which a person experiences a brief incubation period, lasting perhaps only minutes, followed by a sense of overwhelming affect, a sudden inexplicable murder, and perhaps a subsequent amnesia.56 More recent authors have viewed catathymic homicides as occurring within a borderline or psychotic personality organization.57 Homicidal violence is then conceptualized as being perpetrated within the psychopathology of attachment and object relationship theory. The central defense is seen as projective identification through which the perpetrator attributes increasingly malevolent characteristics to another person, who may then become the victim. Although the theory is illustrated by way of erotomania and borderline personalities, the rupture of attachments is often one of the key factors operating in the psychopathology of the severely depressed person. Such a person may be struggling to maintain attachments by delusional formation about his or her worthlessness until he or she becomes overwhelmed by anger and rage.

Final Pathways to Homicide When there is an effort to punish others as well as to be punished, the delusion has come full circle. The incubation period may be the period in which the psychotically depressed and delusional individual awaits a fateful punishment; when it does not occur, things seem confused and intolerable. A prolonged stalemate may lead to provocative behavior and eventually to desperate measures to “end it all.” In one case, a psychotically depressed man performed a truly heroic act, expecting to die because of it. No one who witnessed his behavior was aware that he viewed this act as an opportunity to end his misery after months of suspenseful waiting for that end to come. Similar beliefs that the end is near and that everything is doomed—nihilistic delusions—are often the background for some do-or-die event. States of agitation in depressed individuals or excitement in manic individuals indicate the switch to a state of action from the state of inaction they have been enduring for a prolonged period of time. Mass homicidal violence may be witnessed when people in states of severe depression join cults or movements. Although being indoctrinated into a cult may help a person put depression on hold or alleviate its extreme symptoms, such indoctrination is also likely to leave the person with altered integrative functions of identity, memory, or consciousness.58 In such a dissociated state, those in a group setting are more vulnerable to carrying out acts of suicide or homicide. It is possi-

258


ble that the Reverend James Jones and 990 of his followers, who committed mass suicide in Guyana, were in this category. The final suicidalhomicidal behavior of members of the religious cult known as the Branch Davidians, who died in a conflagration following a standoff with the Bureau of Alcohol, Tobacco and Firearms and the Federal Bureau of Investigation in Waco, TX, in 1993, may have had similar themes. The severely depressed person’s posited dependence on others is congruent with sociological formulations about why certain people become more obedient and conforming than others. In social impact theory, people are seen as becoming more vulnerable depending on the forces that impinge on them in terms of strength or immediacy or in the number of social influences.59 The attraction of the depressed person to strong authority figures as a way out of his or her dilemma is congruent with current knowledge about social impact theory as well as with how severely depressed people grope for relief. This theory accounts for why some people follow a leader to the point of suicide, homicide, or other self-destructive actions. The political implications of such behavior are significant. It is in the midst of extreme forms of delusional self-abasement in psychotic depressions that dissociative states occur. These emerge by way of psychotically depressed individuals distancing themselves from reality in addition to experiencing a severe degree of guilt. To live for any length of time with such a burden lends a sense of unreality as well as otherworldliness. One individual who murdered his wife and children, but failed in a suicide attempt, said that he felt he was already dead before carrying out the homicides and attempting suicide. Performing the acts did not seem to matter. Preceding the homicide, psychotically depressed individuals may have a sense that things are changing in some indefinable way. Another possibility is that these individuals experience Capgras’ syndrome, in which they believe that others they know seem altered, and they experience the delusion that the people in their lives are these altered doubles. If their thinking takes a turn that someone is exploiting them by these phenomena, the potential for violence is increased. Multiple acts of homicide have been committed under the influence of the Capgras’ syndrome. Sometimes an entire family is killed because the individual has become convinced that these are not his or her real family members but only “look-alikes” who have been planted. The impetus to a homicide is when the person expands the belief into an idea that some type of conspiracy exists and that he or she is in danger and must carry out an act viewed as self-protection. The phenomenon of depersonalization or derealization interferes with adequate reality testing. Through these distancing processes, the


259

possibility of homicidal behavior is heightened. Vegetative signs and symptoms that accompany a psychotic depression contribute to states of dissociation by virtue of the actual physical changes that occur. Somatic delusions that take the form of bodily wasting, brains rotting, bowels filled with cement, and so on can accompany physical changes. Another phenomenon noted in some cases has been the temporary loss of hearing that occurs preceding and during the actual perpetration of a homicide in a psychotically depressed person (C.P. Malmquist and P.E. Meehl, unpublished manuscript, July 1988). Again, if these changes are interpreted by the person as meaning that he or she is already dead, then the inhibition against carrying out an act of killing is nullified. It is a matter of ultimate nihilism in which someone already dead kills because nothing matters anymore.

FUTURE RESEARCH PROBLEMS Greater Specificity in Linking Depression and Homicidal Behavior There is a lack of specificity in attempting to determine whether there is any relationship between depression and homicidal behavior. Is there any specific, predisposing factor in a depressed person’s background that helps to predict whether that person will become homicidal? Hypotheses can only be offered based on retrospective examinations of individuals who have committed homicides. It is possible that a depressed person’s vulnerability to become homicidal rests in the same factors that make him or her prone to a psychotic depression. Although this possibility does not answer why some become homicidal, it does suggest an explanation for why such an outcome occurs. In addition to a biological predisposition, there are often events that precipitate a psychotic state. Some events may be related to disappointments, losses, or disillusionments. What to others might be an adverse event is experienced by individuals prone to a psychotic depression as overwhelming. The intensity and depth of their responses go to an extreme, and they exhibit a striking lack of resilience to shift into another mode of thinking. The question is why. Perhaps their brooding and obsessional quality leads them to see others as not “coming through” in meeting their needs. High levels of drive and ambition may have kept a psychotic regression at bay or fended off extreme biological mood swings, but it is difficult to do this repeatedly over a prolonged period of time. The mental state of some—upon realizing that they have committed a homicide—is significant. It reveals not only self-punitive processes,

260


such as the guilt of a murderer before the act, but the dangerous affective state accompanying such distorted thinking. In turn, such a state attenuates the person’s internal signals telling him or her to operate in a contrary way. One research methodology would be to use this clinical insight in either a retrospective or follow-up examination of individuals diagnosed as psychotically depressed to confirm if such a sign portends homicidal violence. If so, this potential indicator would have preventive significance for clinicians working with such individuals who have not yet committed a violent act. Further regression in the affective state brings these individuals to a homicidal level in which distinction between the murderer and the murdered has lost its meaning. This regressed affective state would also correlate with a theoretical position that the special vulnerability of psychotically depressed individuals toward homicidal behavior lies in their cognitive dysfunctioning.60 The answer to whether the cognitive dysfunctioning antedates the onset of the depression, as Beck60 would have it, need not be present to explain how a homicide can occur in some cases. Extreme forms of denial and distortion of reality testing operate in these individuals. Delusional thinking predicated on denial may be elusive to detect. If a homicide occurs, several people will testify that the person acted normally in their view. Perhaps one of the keys in the progression of psychotic depression to the level of homicide is not only a weakening in the perpetrators’ general reality assessment but also the beginning of misinterpreting interactions. As an accompaniment, they possess fewer devices to deal with their hostility because of their tendency to brood about their own deficiencies and worthlessness. Clinical research about transition states into these persons’ evaluation of the seeming worthlessness of everything around them would be quite valuable. If paranoid projections occur, a further element of unpredictability is raised. On one hand, these individuals’ focusing on an external enemy may forestall the ultimate need to destroy themselves and others; on the other hand, it may mobilize them to a homicidal act.

Factors Triggering the Decision to Act In many cases, the reasons why some psychotic depressive individuals commit homicides whereas others simply stay depressed for long periods of time or else commit suicide are impossible to differentiate. From clinical observations, it appears that psychotically depressed people detest themselves to an inordinate degree and hold cognitive distortions about themselves and others. On an operational level, it is not surprising that given their degree of hatred, some of these people can kill oth-


261

ers as well as themselves. Yet psychiatrists’ ability to predict homicide is as limited as it is for predicting suicide. At best, we seem limited to doing risk assessment with weak predictive statements. Although individuals who have engaged in murderous behavior have often castigated themselves to an extraordinary degree, they have not directed this behavior at others to the same degree. The disproportion between their self-hatred and their expressed hatred toward others remains unexplained. Whereas such delusions that are based on self-hatred might explain this degree of self-condemning as well as self-punitive behavior, they do not explain homicidal behavior. The killing of a loved one could be understood in part by saying that the perpetrator’s identity had become so fused with an intimate that killing the intimate and himself or herself was simply one integral act. The common example of this explanation would be when a mother kills her infant and perhaps herself in the midst of a postpartum psychotic depression. In these cases, the mother talks about her love sparing the infant from having to live in a terrible world. A psychobiological fusion has destroyed herself, or a recent part of herself, based on a delusional conviction. Yet most homicides in a psychotically depressed group are not infanticides. What emerges in the larger group of homicides committed by psychotically depressed individuals is the prominence of blaming others. Disappointments in the actual world, or disillusionments in terms of their internalized goals, are not merely experienced as one event among several but viewed with a sense of finality. The author of one study in which five murderers were assessed, with four of them being psychotic, found that the murder was triggered by a sequence of events, beginning with a blow to the perpetrators’ pride, which then tapped into their enormous self-hatred and eventually led to desperate measures—committing the murder—to shore up an unstable pride system.61 The perpetrators had reached a state in which nothing could be regained or righted anymore. This state is similar to unrequited grief that is perceived as having no end-point; the person’s ability to begin the painful process of grieving is either temporarily stymied or was deficient from the beginning. A crucial part of the transition to the homicidal state is the emergence of the idea that somebody must pay for one’s suffering. This is different from the situation of a suicide, in which only relatives and friends may suffer and bear guilt. In some cases, “the world” must pay, which seems to be the mechanism operating in mass killings of loved ones or others. This outcome seems to take place after a prolonged period of regression into a deep and unremittingly depressed state. The validation of delusions in a severely depressed person and the possibility that the delusions are the key variable to homicidal decisions are often viewed skeptically by

262


psychiatrists and certainly by some involved in the legal process. This skepticism merits discussion: Self-derogatory thoughts in seriously depressed people meet the criteria of a fixed belief that is objectively false. To hold to a belief about oneself that is so negative, in the absence of strong, contrary evidence, does not seem any less delusional than to maintain a belief that one is being poisoned or the grandiose idea that one may be some special person. A depressed person’s somatic delusion that his or her body is rotting is somehow more readily accepted as delusional than a psychological belief about the self that he or she is so worthless as to be beyond redemption. These dilemmas are illustrated in the following case.

Case Example A 30-year-old man who showed symptoms of a major depression brooded about suicide for months. His usual pattern of a high level of work performance had fallen off. In an attempt to lessen his depressed state, he had an affair outside his marriage, which only compounded his guilt. After months of ambivalent procrastinating about his state, and in the midst of watching the television movie Barabbas, in which a guiltridden man seeks redemption and sets Rome on fire, the 30-year-old man committed multiple homicides. Near the end of the movie, he shot his wife, who was sitting on a couch opposite him, in the forehead and then ran around the house spilling gasoline and setting the house on fire. After all four of his children burned to death, he proceeded to tie a gun with a string attached to the trigger to a stove in the garage. He then shot himself at a distance of 8 feet. That bullet went through the right side of the sternum. He then stepped to the right and shot himself a second time in an attempt to have the bullet enter near his heart. The second bullet ricocheted around his rib cage and exited. He then slipped a rope around his neck and wrists and hung himself, with his wrists extended through the loops on an inverted-U iron clothes pole. Some teenagers who were attracted by the fire discovered him, and he survived. The jury found him guilty of second-degree murder despite expert testimony regarding his insanity defense.62

Resistance to Accepting Delusionality in Depressed People There are reasons why it is difficult to accept the reality of psychotic thinking in severely depressed people. Even the seriously depressed person does not appear that different from others in contrast to those with some other types of psychosis. If others view such a person as psychotic, it raises implications for others, including clinicians themselves, who must deal with such disturbed states. Second, a tenuous link exists between verbalizations made by psychotically depressed individuals and the degree to which it is believed that their words parallel their actual in-


263

ternalized mental states. Pleas for help, which clinicians hear endlessly from delusional people, have a side effect of increasing clinicians’ skepticism about the validity of self-abasement claims. Thus, the distinction is blurred between the “worried well” and the legitimate cases of those who are severely depressed. Demanding, complaining, or provocative aspects complicate matters further because they convey that the person is complaining about someone other than himself or herself. However, even if these self-condemnatory accusations are taken at face value, a third difficulty involves semantic interpretation. A person’s self-assessment often lies somewhere between the realm of fact and his or her value system. If the person’s statements are seen as more reflective of his or her value system and not a factual state, then the statements tend to be discredited or overlooked for possibly being delusional. If the clinician can determine where such statements can be referenced and checked against facts, there is a greater tendency to accept them as delusional. For example, the statement made by someone that he or she is at fault for a tornado that has caused damage and some deaths is likely to sensitize the clinician to the possibility of a delusion, whereas a person who simply states, “I’m a worthless person” may not. The problem of different normative values raises other issues. If people with strong moral beliefs about a matter violate these beliefs, it is difficult to say that a subsequent guilt-ridden attitude is delusional. For example, someone who believes that engaging in homosexual behavior is morally wrong but who then acts in that manner and begins to brood and condemn himself or herself would not automatically be classified as delusional. What is needed is an appraisal of the severity of the self-condemnation, its persistence, and its amenability to change. In the psychiatrist’s efforts to infer that reality testing has become significantly impaired, the confusion resides in value statements where an objective reality is elusive. Such confusion leads to a related difficulty. Even if the value system of an individual can be explained, at least in philosophical terms, it is still difficult to attain a sense of agreement on objective matters. This holds not only among clinicians but also for individual citizens called on to make judgments about guilt in a public forum such as a jury, which involves the truth status of a person’s self-appraisal and subsequent responsibility for behavior in a criminal court.

Civil Liability A different type of problem is increasing in our litigious society. Although the clinician may find it difficult enough to diagnose psychotic

264


depressions or emerging states of manic psychoses—and the conditions do tend to be underdiagnosed—there are increasing practical implications for failure to make such diagnoses. Clinical and research limitations may not catch up to demands for accountability by relatives in distressed states or relatives of innocent victims. Civil lawsuits involving medical liability in connection with acts of violence are increasing. These lawsuits arise not only when such a diagnosis is missed but also when its severity is misgauged. Jury awards may be horrendous, even in commonplace clinical situations; for example, a $5.8 million jury verdict was awarded in the case of a man who told the police he was depressed and suicidal subsequent to breaking up with his girlfriend. He was involuntarily committed to a state hospital for 5 days, then discharged on the belief that he could no longer be kept hospitalized. Two weeks after being released for outpatient treatment, he entered a house with a .357 Magnum and shot three people, killing one and wounding the other two; afterward, the man committed suicide. The claim of the plaintiffs was based on a failure to diagnose a major depression with its implications for violent conduct. In such cases, a documented risk assessment before discharge is most valuable in demonstrating that good professional judgment was exercised. If questions are present of continuing dangerousness connected to a mental illness, issues of a duty to warn or protect third parties may arise in certain jurisdictions.63

Infanticides and Filicides Although not all acts of young children being killed by parents or stepparents are due to depressive conditions or bipolar states, a significant number of them do fall into this category and thus these types of killings are discussed here. Often, a comorbid diagnosis is present from the realm of substance dependence or a personality disorder. In many cases, some of the diagnostic possibilities discussed in other chapters of this book are relevant and become important in distinguishing postpartum psychoses from other disorders. Distinctions are often made based on the age of the child victim. The term feticide has been used to describe the killing of a fetus; these cases arise when a pregnant woman is killed and the fetus killed along with the mother. The perinatal period is from the 28th week of gestation through the 7th day after birth, and a killing in that period would be called a perinatal death. The neonatal period is the first 28 days of life, and hence a death in that period is a neonatal killing. Usually, the period of infancy is medically thought of as the first year of life, but legal definitions of infancy vary widely, sometimes up


265

to 5 years. A killing in this period is then an infanticide. The term filicide is used for killings beyond infancy, but generically it covers the murder of a child irrespective of age. Apart from homicidal acts, neonatal mortality is usually associated with diseases of preterm birth, congenital anomalies, and low birth weight. Some interesting facts are that in the United States there are annually 6 million pregnancies and 4 million live births, with 18,000 neonatal deaths and 28,000 infant deaths.64 Twelve percent of births are to teenage females between 15 and 19 years of age. Historically, infanticide was one of the practices employed for regulating the number of children in a family due to economic pressures or excess numbers of offspring. As late as the early 1800s in Europe, up to a third of live-born infants were killed or abandoned by parents.65 In the United States, homicide is now the leading cause of infant deaths due to injury, with more than 80% of infant homicides considered fatal child abuse. The highest-risk group is composed of the second or subsequent infants born to a mother who is younger than 17 years of age.66 Classification systems regarding infanticide have varied. Some have simply divided the deaths into maternal or paternal filicide, and other systems have derived their own categories. Resnick’s67 1970 article used a motivational system: altruistic acts; psychotic; unwanted child; revenge on a spouse; and accidental. Another typology of filicide suggested five categories: pathological filicide that included altruistic killings where the mother was diagnosed with major depression with psychotic features; accidental filicide connected to physical abuse; retaliating filicide for revenge against a spouse or lover; neonaticide of an unwanted child; and paternal filicide.68 More recent attempts at classification have focused on biological factors and diagnoses. Neurohumoral factors contributing to an affective disorder in pregnancy and in the postpartum period indicate the sensitivity of women to the acute withdrawal effects of estrogen and possibly progesterone in the postpartum period.69 Noting that postpartum psychoses are not a distinct diagnostic entity, despite depressive disorders being elevated in the first postpartum weeks, Kohl70 discussed whether the condition is closer to schizophrenia or the affective spectrum. Her suggestion was that the specifier postpartum psychosis be attached to whatever mental illness is present. Others see more of a link with bipolar disorders.71 Another study reviewed 55 filicidal women over a 22-year time period who were being evaluated for criminal responsibility or competence.72 At the time of the filicide, 52.7% had psychotic symptoms and were more likely to have a history of substance abuse.

266


Hogg73 reviewed 22 instances of “murdering mothers” in Toronto over a 60-year period. A repetitive pattern was noted in that the women all seemed lucid and “normal” when seen within a few hours of the killings. They were usually described in reports by police and professionals as showing little crying and offering no denial of the acts and were viewed as cold-hearted, normal, and free of mental illness. These women were quickly booked and charged with murder. However, within 24 hours, most of them showed signs of overt illness of a psychotic depressive nature and belatedly attempted suicide. The time right after the killings was seen as a lucid interval occurring in the midst of being psychotically depressed, and the act of killing had temporarily shocked them out of it. Spinelli74 evaluated 17 women to determine their mental state at the time of the alleged offense (16 neonaticides and 1 attempted act). She concluded that there was a denial of pregnancy in the women, who also shared other characteristics such as unassisted deliveries, dissociative hallucinations, and intermittent amnesia at delivery. She referred to a spectrum of disavowal involving either a blatant denial of the pregnancy or only an intermittent awareness of an intolerable reality. Her conclusion was that the women had not been able to maintain a successful dissociative response, which induced a dissociative psychosis (“hysterical psychosis”) in which the experience was like a waking dream for them. These various formulations may all be played out at trial when a woman is indicted for an act of infanticide. The focus is on female perpetrators because males who kill an infant seem to be operating with motives of anger, rage, or being excessively burdened by crisis situations, such as economic problems or a physically handicapped child. In situations involving male perpetrators, the customary criminal procedures prevail. There have been several high-publicity cases of women who killed their children. In these cases, a mental illness defense may be raised, but in my experience an insanity defense is rarely successful. This outcome may seem surprising, because the immediate public response is some variation on the theme that “a woman who kills her child must be crazy.” An initial issue is always whether a mental disorder is present, and if so, what diagnosis. DSM-IV-TR does not give a separate diagnosis for postpartum disorders but rather provides that the specifier “with postpartum onset” be attached to one of the mood disorder diagnoses if the onset is within 4 weeks after childbirth. This somewhat arbitrary approach may thus overlook many cases in which either the onset has been more gradual or the symptoms have been denied or concealed. In many cases of


267

postpartum psychosis, the fluctuating state of signs and symptoms further confuses the question regarding the presence of a serious mental illness. This becomes particularly important when the scene shifts to courtrooms, because the gradual unfolding of a psychosis may have occurred over several months. When proceedings are in the adversarial setting of legal proceedings, various defenses may be raised, such as insanity, diminished capacity, or an involuntary act. Because usually no specific statutes separate infanticidal homicidal acts from other homicides, the result is a wide variance in the defenses and outcomes of cases. High publicity in a given case may also play a contributing role in determining how the matter is handled within a given jurisdiction. Surprisingly, neonaticide has produced few appellate decisions despite its regular occurrence. This might be related to the bodies of newborns not being discovered or the difficulties in attributing the act to a particular person when the pregnancy was hidden, so that the identity of the mother of a dead or discarded newborn is not known in many cases.

Case Examples A well-educated professional woman had married later in life to allow herself to pursue a career. An earlier therapeutic abortion had taken place so that she could finish her education. Later, she married the father and was pregnant with their desired child. Unfortunately, the infant was born with several congenital anomalies. Ongoing care was required, and at 6 months the infant was still being fed via a feeding tube. About 3 months after delivery, ideas slowly emerged that her producing a “defective child” was the result of some failure on her part. About 1 month later, she developed the idea that she owed it to her husband to kill the infant, even though she insisted that she loved the child and her care had been exemplary. These ideas preoccupied her for 2 months. One evening she decided she had to carry out these ideas and would do so the next morning. She slept peacefully during the night with the reassurance she had solved the problem. When she awoke, she got a kitchen knife and plunged it into the infant’s heart. Thinking through the consequences of the act had never occurred to her, because resolution of the problem preempted all thoughts. She later suicided while awaiting trial. The Andrea Yates case in Texas in 2003 involved the drowning of her five children, ages 6 months to 7 years.75 Andrea had been a high school valedictorian and a nurse. Her history of psychiatric illness involved feeling Satan’s presence and hearing him. After her oldest child’s birth in 1994, Satan’s voice told her to stab him with a knife, but she told no one for fear that Satan would harm her or the children. After the birth of her fifth child, she began to appear “catatonic.” There were two psychiatric hospitalizations, but she continued to deteriorate. Two weeks be-

268


fore the final events, her antipsychotic medication was stopped, and acute psychotic symptoms emerged, with cartoon characters calling her a bad mother and command hallucinations saying to kill the children to save them from going to hell. She was charged with capital murder, which carried the death penalty, and in 3½ hours the jury returned a guilty verdict. The prosecution then requested the jury to vote on the capital punishment issue, but they voted for life imprisonment. On appeal a new trial was ordered.76 The reversal was based on the “false testimony” given by the prosecution’s psychiatrist, Dr. Park Dietz. During the trial, five mental health experts had testified that the defendant did not know right from wrong or that she thought what she did was right. Dr. Dietz had testified that she knew right from wrong. His reasoning was that if her thoughts were coming from Satan, she must have known they were wrong. Furthermore, if she believed she was saving the children, she would have shared her plan with others and not hidden them as she did. If she believed Satan was going to harm the children, would she not have called the police or a pastor or sent the children away? Because four of the five children were found soaking wet and covered with a sheet on the mother’s bed, the conclusion was that they were covered out of guilt or shame about what she had done. However, the key to reversing the conviction and ordering a new trial was Dr. Dietz’s testimony that a few weeks before the homicides, an episode of the television program Law and Order had been shown that involved a woman with a postpartum depression who drowned her children in a bathtub and was found insane. When checking this testimony, attorneys discovered that no such episode had been aired on television. The false testimony was discovered after the jury verdict but before sentencing. The defense moved for a mistrial, but the court rejected the motion and instead allowed a stipulation that Dr. Dietz’s testimony was in error regarding the Law and Order program. Because the false testimony was seen as material to the jury’s verdict, the case was reversed and remanded by the appellate court.

When children are past infancy, the postpartum attribution becomes more tenuous. We then enter the realm of either 1) more mundane motives operating, such as economic factors, seeking sympathy, or revenge—standard motives operating in many homicides; or 2) certain types of psychological defenses related to mentally disturbed thinking. The drama Medea by Euripides, written in the fifth century B .C., describes Medea’s response to her husband, Jason, abandoning her for another woman. In response Medea feels insulted but also reduced in stature as a human being. Her sense of injury is that she experienced the abandonment as a public humiliation. The sense of being wronged produced a sense of righteousness about the degree of her hate for Jason that also allowed her to feel justified in killing their two sons. The feeling of being publicly mocked prevailed and justified her going from a passive state of feeling wronged to reversing her humiliated state.


269

However, the price was that of murdering her two sons and suffering whatever the social consequences might be.77 Diverse socioeconomic and racial issues may intrude in some cases. The Susan Smith case captured the public imagination because it received widespread coverage. Her first version of events was that a black man had forced her at gunpoint to relinquish her red Mazda along with her two sons, ages 3 years and 14 months, who were in the car. She then began to appear on local and national television for 9 days, pleading for their return. Then, just hours after appearing on such programs, she confessed to strapping the boys into the car and rolling it into a nearby lake. The boys had been strapped into children’s car seats and the windows rolled up, so a slow death would have occurred. A diver reported he saw a little hand pressed against the glass. The public felt as badly duped as the news media. The image reversal was from an assessed role of a mother who was a beleaguered angel to that of some type of monster.78

Case Example A black mother, age 24 years, was charged with murder and attempted murder after she jumped from a bridge into a river with her two children. She had had several earlier hospitalizations with diagnoses usually of bipolar disorder with psychotic features. On those admissions, she was usually agitated and hallucinating. The admissions were often precipitated by bizarre behavior, such as taking her four children out at 3:00 or 4:00 in the morning and roaming the streets, singing loudly to keep Satan away. She would then be placed on medication and discharged after a few days with instructions to stay on medications, which she would do for a few months and then stop. She saw herself as pursuing a singing career while living in subsidized public housing with the children. In more recent depressive episodes, she developed the idea that she was the “Black Messiah.” Another delusional belief was that “those in power,” such as the Central Intelligence Agency or the Masons, were trying to thwart her ambitions, because she believed they were spying on her from adjacent apartment buildings and bugging her phone. On the day of the homicide, she believed a terrible catastrophe was going to occur in which “good people” would be destroyed. She unsuccessfully looked for some sign of hope and believed this was her last day of life on earth. Her idea was that she had to give a message of the impending disaster to the good people to warn them. While pushing the two infants in a stroller on the bridge, she thought there might be one last option to see whether there was any hope. She decided to throw one of the children up in the air over the river while standing on the bridge. If the child “ascended” there was hope they could all be saved. When the child did not ascend, she kissed the other child as she had the first and also threw him over the bridge. Before jumping herself, she yelled, “Freedom.” One infant died but she

270


and the other somehow survived the jump. One viewpoint was that this was an intentional political act and therefore met legal criteria for murder. Another viewpoint was that she was acting on delusional beliefs that the acts were not wrong and thus would meet an insanity test. A reduced murder plea was negotiated.

These types of cases involve broader issues than postpartum depressive psychoses. They are filicidal acts that raise the panoply of approaches to take in terms of how they are handled in different legal jurisdictions. One approach would be a straight type of legal charge for some degree of homicide in which plea negotiations would follow for the degree of culpability that would be seen as just. This approach is the most common outcome in these cases. It saves the prosecution from the possibility of losing an insanity verdict but also allows the defense to feel vindicated by a reduced plea. The ultimate legal question of insanity is then bypassed even though it could be raised in many of these cases. Aside from insanity, another option the defense considers is a diminished-responsibility defense if a jurisdiction allows it. A variation is to get into the irresistible-impulse question, such as claiming the act was involuntary, like an automatism.79 The argument would be that the parent committing the act, as in the case examples, should be excused because the act is connected to a mental disability and is not the product of the person’s efforts. The analogy is to a state of sleepwalking violence in a depersonalization disorder, especially if a neonaticidal death occurs.80 The difficulty with this option is that it raises all the joint unresolved legal and medical issues involving lack of capacity to control one’s behavior as well as violence occurring during states of impaired consciousness. Perhaps the endless legal battles over these types of cases are more typical in the American criminal justice system than in other countries. For over 50 years in England, no woman found guilty of infanticide has been incarcerated.81 These women are either put on probation with accompanying treatment or placed in psychiatric hospitals.

Iatrogenic Homicides: The Selective Serotonin Reuptake Inhibitor Problem A good deal of discussion and debate has arisen over the consequences of antidepressant use involving the selective serotonin reuptake inhibitor (SSRI) class. Apart from their benefit to many depressed patients, a primary issue has been whether these drugs, or some of them, increase the probabilities for violent behavior. Previous discussions focused on the excessive prescribing of this class of drugs or their use to enhance creativity or well-being in nondepressed people. Continuing cultural


271

and moral arguments about the use of SSRIs are not pursued here beyond noting the concern some have about the United States as a “Prozac nation.”82 Two sets of questions are germane to the issue of violence and SSRI medications. One is an unresolved scientific question regarding the actual determination of whether these drugs increase the possibility of violence in some people. If so, the question is which people are affected— raising inquiries specifically about adolescents, although the question is not restricted to this one age group. In turn, questions then follow about possible differences in response based on sex, ethnicity, period of time on the drug, and the specific SSRI taken. The second set of questions arises in legal settings. The defense in a homicide case may raise a question of whether an SSRI played a primary causal role in inducing homicidal behavior. Civil cases may proceed from relatives of a person who suicided on the basis of the SSRI contributing to this kind of violence. Relatives of a person murdered may also bring a suit not only against the person perpetrating a homicide but also the pharmaceutical manufacturer. Well-designed clinical studies illustrate the continued difficulty in drawing definitive conclusions about the enhanced possibility of violence and SSRI use. Although many articles examine the side effects of SSRI drugs, there are conflicting reports regarding violence potential. Those studies that do investigate the potential for violence focus on suicide rather than homicide because of the difficulty in studying homicidal potential before such an act. A United Kingdom study used the U.K. General Practice Research Data Base for 1993–1999 based on 159,810 users of four antidepressant drugs: amitriptyline, fluoxetine, paroxetine, and dothiepin.83 In the study, a person could be taking only one of these drugs within 90 days before the onset of suicidal behavior or ideation. The risk of suicidal behavior after starting antidepressant treatment was similar among users of the first three drugs compared with dothiepin. An increased relative risk was seen in 555 newly diagnosed cases compared with 2,062 control subjects in the first month of starting antidepressants, especially in the first 1–9 days. The authors noted a possible small increase in risk among those starting paroxetine but thought the risk could be confounded by the severity of the depression. The conclusion was that no substantial difference in effect was found from the four drugs on people ages 10–19 years. Beginning in 1990, anecdotal and individual case reports about fluoxetine and suicidality began to appear. The other SSRIs were included in the debate as they were marketed. Healy84 became an advocate for the view that SSRIs carry a suicidogenic potential. A review of the epi-

272


demiological literature of Western populations and clinical trial databases from several countries did not find support for SSRIs increasing suicidality in patients being treated for depression or in those taking the medications for some other reason.85 Another group then explored the issue by a review of randomized controlled studies, meta-analyses of clinical trials, and epidemiological studies.86 The meta-analyses revealed an excess of suicidal acts for those in active treatment compared with those given a placebo (2.4 to 1.0). These authors concluded the null hypothesis was difficult to sustain. However, another study assessed “completed suicides” among patients assigned either to an SSRI, some other antidepressant, or a placebo and found no difference in suicide risk among the three groups.87 Again, Healy’s review88 of randomized clinical trials and meta-analyses concluded that the data indicated a possible doubling of the relative risk for suicides and suicide attempts compared with older antidepressants or no treatment. Several articles deal with hypotheses of what may mediate the violence connected with serotonin. Many of the hypotheses deal with aggressive and impulsive behavior related to low serotonergic activity and high plasma levels of testosterone; they state that the combination affects neural mechanisms involved in the expression of aggressive behaviors.89 Serotonin is seen as playing a role in inhibition of impulses and regulation of emotional control.90 Dysfunction in the serotonergic system is perceived as trait dependent and associated with disturbances in regulation of anxiety, impulsivity, and aggression.91 One study focused on 12 males with a history of criminal behavior; a placebo was given to 6 of them and a placebo and paroxetine to the other 6.92 At the end of 21 days, laboratory testing showed a significant decrease in impulsive and aggressive responses in subjects receiving paroxetine (20 mg/day). From a totally different perspective, a psychoanalytic article with commentaries ascribed the function of SSRIs as attenuating internally directed aggression as well as contributing to the disappearance of externally directed aggression.93,94 The authors did not address additional questions raised about the possibility of increased suicidality and homicidality in some people. These conflicting results and viewpoints have produced a continuing debate about both children and adults and related suicidality and homicidality. The uncertainty about the safety of using certain SSRIs led to a report in 2003 from the British Medicines and Healthcare Products Regulatory Agency; the report recommended cessation of the use of all antidepressants except fluoxetine in depressed youths under age 18. The American response to the report was that the definition of “suicidality” was unclear because it included ideation, self-harm, and actual suicide


273

attempts.95 Drug regulators in Britain strongly urged doctors not to use certain antidepressants (Paxil, Luvox, and Zoloft) for childhood depressions, whereas in the United States the Food and Drug Administration (FDA), relying on the same data, did not make a decision on the drugs’ safety.96 Another approach has been that of the New York attorney general, who brought a lawsuit against the manufacturer, Glaxo SmithKline, charging the company with fraud in concealing negative information about paroxetine.97 These ongoing scientific and political debates about the safety of certain antidepressants cannot be tested in a similar manner for homicidal behavior, yet they are relevant. For extreme violence, an after-theact evaluation would be necessary, because violent fantasies could only tell us whether the fantasies increased—or not—following the use of an SSRI. However, fantasies are a long way from a homicide. Despite these unresolved scientific questions, SSRI issues have figured in both the civil and criminal legal arenas. A civil case may be brought against physicians as well as the pharmaceutical firm when someone on an SSRI commits suicide or attempts a suicide. Surviving relatives may bring the action for a wrongful death or a products liability allegation. In criminal actions, the perpetrator’s defense may include arguments that SSRIs induce agitated or out-of-control states for homicidal violence. Some illustrative cases follow. 1. By the early 1990s, the “Prozac defense” had been tried in some criminal cases. It was argued that the drug spurred patients to commit violent acts. By 1991, such a defense had been tried unsuccessfully in 17 criminal cases.98 In a study of 20 men arrested on murder charges, their whole blood serotonin level was found elevated compared with a nonviolent reference group.99 This result was noted with other studies in which violent behavior was associated with alterations in whole blood serotonin level. 2. The “failure to warn” claim against the drug companies has been raised in several cases. In the Motus v. Pfizer case, a man suicided 6 days after Zoloft was prescribed. His widow brought suit against Pfizer, the pharmaceutical company, for failure to warn about the drug’s suicidal possibility.100 The physician-prescriber indicated that he would have prescribed the medication even with adequate warnings. California law required proof that Pfizer’s failure to warn was a “substantial factor” in bringing about the suicide. The burden was on the plaintiff to demonstrate that the undisclosed risk was sufficiently high that it would have changed the treating physician’s decision to prescribe. The suit was dismissed, a decision that was upheld on appeal.

274


3. Pfizer was again the defendant in a case in which a 13-year-old boy committed suicide 6 days after Zoloft was prescribed.101 Liability was alleged for marketing defects—misrepresentation and negligent failure to warn about the dangers of drug-induced suicide. A federal court ruled the manufacturer was not liable when there was no evidence of fraud or a failure to warn. The testimony of a medical expert was disallowed under Daubert as not following accepted methodology in testing the hypothesis that Zoloft causes suicide. Summary judgment for the defense was upheld on appeal. The district court appointed independent specialists to evaluate the expert’s opinions, and they concluded the opinion of the plaintiff’s expert was unscientific. The trial court had allowed the expert to express revised opinions but noted that the calculations of the risk of suicide had not been subjected to peer review, that there had been no placebo control, that a small sample had been used, and that there had been interactions between the participants and the researchers. 4. The estate of a family brought suit in Tobin v. SmithKline Beecham after a 60-year-old depressed man shot and killed his wife, 31-year-old daughter, 9-month-old granddaughter, and then himself.102 Three days before these events, he had obtained prescriptions for paroxetine and zolpidem for sleep. Five prior episodes of depression had kept him out of work, and he had a sporadic history of taking medications. The drug company initially tried a motion for summary judgment on the basis of the “learned intermediary doctrine,” in which physicians have a duty to be knowledgeable about prescribed medications. The motion was denied because the physician stated that warnings about the drug might have changed his prescribing decision, contrary to the case discussed earlier. The plaintiff’s claim was based on negligence for failure to test paroxetine for safety and inadequate warnings, whereas the defense claimed the suicide was the result of being depressed and that the medication was not the proximate cause of death. In this case, the opinion of the expert was allowed. The expert testified that deliberate self-harm for people taking SSRIs is five and onehalf times higher than for someone taking tricyclic antidepressants, and a study in support of the opinion was cited.103 The Tobin case was the first paroxetine wrongful death/suicide case to go to a jury trial in the United States, and the jury awarded $6 million to the plaintiffs.104 The verdict was initially appealed but later settled and dismissed. 5. Another paroxetine case was brought by the same attorney as in the Tobin case and using the same expert witness, Dr. Healy. An individual had died by carbon monoxide poisoning after beginning an


275

increased dosage. The plaintiffs sought to bar the defense from deposing Dr. Healy on the basis of collateral estoppel.105 This motion meant that the court would decide that the issue in this case was identical with the one previously decided and that the prior action was decided on its merits. The court denied this motion because the judge in the previous case had not determined whether the manufacturer was at fault for failing to test, failing to warn, or both. “Collateral estoppel is not intended to operate as a shield against scrutiny of the reliability of the underlying science of the experts.”105 The case was settled out of court. The question of causality continues to loom large because expert opinions tend to use a limited pool of data to support causality conclusions. Experts often cite their own publications or that of similar specialists and use studies from the early 1990s and unpublished data of pharmaceutical companies.103 What is needed are new analyses and the release by companies of their various test data, which will hopefully provide more consistent conclusions. When an SSRI type of defense is used in criminal trials, the effort is to obtain either an insanity verdict or one of diminished responsibility. The question becomes whether the SSRIs changed the mental state of the individual at the time of committing the criminal act. It might be argued that but for the medication, the person would not have acted as he or she did. Publicity has focused on the effects of SSRIs on adolescents, but the FDA asked drug manufacturers to put detailed warnings about the possibility of increased suicidal behavior on the labels of 10 antidepressants, for both children and adults.106 So far, the requests have not included warnings about the possibility of homicidal behavior. The Pittman case in South Carolina illustrates many of the issues under discussion.107 Often a host of psychosocial adverse factors complicate this type of case.

Case Example Christopher Pittman was 12 years old when he killed his grandparents with shotgun blasts and then set their home on fire, took their money, and drove away in their car. He stated that he was hearing voices and that “something kept telling me to do it.” He had been taking Zoloft at the time of the killings. Christopher’s mother had left when he was 2 years old. His parents had attempted reconciliation in 2001, the same year Christopher killed his grandparents. When the parents again separated, he threatened to kill himself and was hospitalized as depressed. He was put on paroxe-

276


tine, but after 1 week his father removed him from the hospital and sent him to live with the grandparents, where he was put on Zoloft. After a few weeks, Christopher got into a dispute on the school bus, and his grandparents threatened to send him back to his father. The next day, they were dead. The boy later wrote a letter that his father read at an FDA hearing. It stated, “Through the whole thing, it was like watching your favorite TV show. You know what is going to happen but you can’t do anything to stop it.”

The attorney who had won the Tobin case in Wyoming for $6 million had become part of the boy’s defense team after being contacted by the International Coalition for Drug Awareness, a Utah group opposed to antidepressant use. The group has been involved in several murder cases in which the defendant was on antidepressants. At this point, the issue of SSRIs and violence is unresolved by scientific criteria. Investigators have arrived at different conclusions. The need is to disaggregate issues, such as the specific SSRI being used, suicidal or homicidal behavior, and variables such as the age of the user, dosage of the drug, length of use, role of other drugs being used, and comorbidity. The studies need much more specificity involving these variables. In the meantime, lawsuits are sure to continue in this ambiguous atmosphere.

REFERENCES 1. Goodwin FK, Jamison KR: Manic-Depressive Illness. New York, Oxford University Press, 1990 2. Paykel ES: Handbook of Affective Disorders, 2nd Edition. New York, Guilford, 1992 3. Gotlib IH, Hammen CL (eds): Handbook of Depression. New York, Guilford, 2002 4. Tardiff K: The current state of psychiatry in the treatment of violent patients. Am J Psychiatry 149:493–499, 1992 5. Rosenbaum M, Bennett B: Homicide and depression. Am J Psychiatry 143:367–370, 1986 6. Reiss AJ Jr, Roth JA: Understanding and Preventing Violence. Washington, DC, National Academy Press, 1993 7. Cassano GB, Rucci P, Frank E, et al: The mood spectrum in unipolar and bipolar disorder: arguments for a unitary approach. Am J Psychiatry 161:1264–1269, 2004 8. Grove SM, Andreasen NC: Concepts, diagnosis and classification, in Handbook of Affective Disorders, 2nd Edition. Edited by Paykel ES. New York, Guilford, 1992, pp 25–42


277

9. Keller MD, Lavori PW, Mueller TI, et al: Time to recovery, chronicity, and levels of psychopathology in major depression: a 5-year prospective follow-up of 431 subjects. Arch Gen Psychiatry 49:809–816, 1992 10. Hill SK, Keshavan MS, Thase ME, et al: Neuropsychological dysfunction in antipsychotic-naïve first-episode unipolar psychotic depression. Am J Psychiatry 161:996–1003, 2004 11. Fava M, Rosenbaum JF, Pava JA, et al: Anger attacks in unipolar depression, part 1: clinical correlates and response to fluoxetine treatment. Am J Psychiatry 150:1158–1163, 1993 12. Dougherty DD, Rauch SL, Deckersbach T: Ventromedial prefrontal cortex and amygdala dysfunction during an anger induction positron emission tomography study in patients with major depressive disorder with anger attacks. Arch Gen Psychiatry 61:795–804, 2004 13. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 14. Kendler KS: Mood-incongruent psychotic affective illness. Arch Gen Psychiatry 48:362–369, 1991 15. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC, American Psychiatric Association, 1980 16. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987 17. Black DW, Nasrallah A: Hallucinations and delusions in 1715 patients with unipolar and bipolar affective disorders. Psychopathology 22:28–34, 1989 18. Silverstone T, Hunt N: Symptoms and assessment of mania, in Handbook of Affective Disorders, 2nd Edition. Edited by Paykel ES. New York, Guilford, 1992, pp 15–24 19. Schatzberg AF, Rothschild AJ: Psychotic (delusional) major depression: should it be included as a distinct syndrome in DSM-IV? Am J Psychiatry 149:733–745, 1992 20. Taylor PJ, Mullen P, Wessely S: Psychosis, violence and crime, in Forensic Psychiatry: Clinical, Ethical and Legal Issues. Edited by Gunn J, Taylor PJ. Oxford, UK, Heinemann, 1991, pp 329–372 21. Dubovsky SL, Thomas M: Psychotic depression: advances in conceptualization and treatment. Hosp Community Psychiatry 43:1189–1190, 1992 22. Lykouras E, Christodoulou GN, Malliaras D: Type and content of delusions in unipolar psychotic depression. J Affect Disord 9:249–252, 1987 23. Coryell W, Zimmerman M: Demographic, historical, and symptomatic features of the nonmanic psychoses. J Nerv Ment Dis 174:585–592, 1986 24. Parker G: Classifying depression: should paradigms lost be regained? Am J Psychiatry 157:1195–1203, 2000

278


25. Ohayon MM, Schatzberg AF: Prevalence of depressive episodes with psychotic features in the general population. Am J Psychiatry 159:1855–1861, 2002 26. Coryell W, Leon AC, Turvey C, et al: The significance of psychotic features in manic episodes: a report from the NIMH collaborative study. J Affect Disord 67:79–88, 2001 27. Hodgins S: The criminality of mentally disordered persons, in Mental Disorder and Crime. Edited by Hodgins S. Newbury Park, CA, Sage, 1993, pp 3–21 28. Monahan J: Mental disorder and violent behavior. Am Psychol 47:511–521, 1992 29. Klassen D, O’Connor W: Assessing the risk of violence in released mental patients: a cross-validation study. J Consult Clin Psychol 1:75–81, 1990 30. Teplin L: The prevalence of severe mental disorder among male urban jail detainees: comparison with the Epidemiologic Catchment Area program. Am J Public Health 80:663–669, 1990 31. Lindqvist P: Criminal homicide in northern Sweden 1970–1981: alcohol intoxication, alcohol abuse and mental disease. Int J Law Psychiatry 8:19–37, 1986 32. Gottlieb P, Gabrielson G, Kramp P: Psychotic homicides in Copenhagen from 1959 to 1988. Acta Psychiatr Scand 76:285–292, 1987 33. Swanson JW, Holzer CE III, Ganju VK, et al: Violence and psychiatric disorder in the community: evidence from the Epidemiologic Catchment Area surveys. Hosp Community Psychiatry 41:761–770, 1990 34. Robins LN, Helzer JE, Croughan J, et al: National Institute of Mental Health Diagnostic Interview Schedule: its history, characteristics, and validity. Arch Gen Psychiatry 38:381–389, 1981 35. Dohrenwend BP, Shrout P, Egri G, et al: Measures of nonspecific psychological distress and other dimensions of psychopathology in the general population. Arch Gen Psychiatry 37:1229–1236, 1980 36. Link B, Cullen F, Andrew H: The violent and illegal behavior of mental patients reconsidered. Am Sociol Rev 57:275–292, 1992 37. Rosenbaum M: The role of depression in couples involved in murdersuicide and homicide. Am J Psychiatry 147:1036–1039, 1990 38. Palmer S, Humphrey J: Offender–victim relationships in criminal homicide followed by offender’s suicide, North Carolina, 1972–1977. Suicide Life Threat Behav 10:106–118, 1980 39. Coid J: The epidemiology of abnormal homicide and murder followed by suicide. Psychol Med 13:855–860, 1983 40. Curran PS, Finlay RJ, McGarry PJ: Trends in suicide in Northern Ireland, 1960–1986. Ir J Psychol Med 5:98–102, 1988 41. West DJ: Murder Followed by Suicide. Cambridge, MA, Harvard University Press, 1966 42. Kendell RD: Relationship between aggression and depression. Arch Gen Psychiatry 22:308–318, 1970


279

43. Malmquist CP: Psychiatric aspects of familicide. Bull Am Acad Psychiatry Law 8:298–304, 1981 44. Akiskal HS, Puzantian VR: Psychotic forms of depression and mania. Psychiatr Clin North Am 2:419–439, 1979 45. Kendler KS, Glazer SM, Morgenstern H: Dimensions of delusional experience. Am J Psychiatry 140:466–469, 1983 46. Woodis G: Depression and crime. British Journal of Delinquency 8:85–93, 1957 47. Lamberti J, Blackman N, Weiss J: The sudden murderer: a preliminary report. Journal of Social Therapy 4:2–10, 1958 48. Weiss J, Lamberti J, Blackman N: The sudden murderer: a comparative analysis. Arch Gen Psychiatry 2:669–678, 1960 49. Blackman N, Weiss J, Lamberti J: The sudden murderer, III: clues to preventive interaction. Arch Gen Psychiatry 8:289–294, 1963 50. American Psychiatric Association: Diagnostic and Statistical Manual: Mental Disorders. Washington, DC, American Psychiatric Association, 1952 51. Wertham F: The catathymic crisis. Arch Neurol Psychiatry 37:974–978, 1937 52. Hinsie LE, Campbell RJ: Psychiatric Dictionary, 3rd Edition. New York, Oxford University Press, 1960, p 111 53. Kraepelin E: Lectures on Clinical Psychiatry, 3rd English Edition. Edited by Johnstone T. New York, W Wood, 1913 54. Wertham F: The Sign for Cain. New York, Macmillan, 1966 55. Revitch E, Schlesinger L: Psychopathology of Homicide. Springfield, IL, Charles C Thomas, 1981 56. Meloy JR: Violent Attachments. Northvale, NJ, Jason Aronson, 1992 57. West LJ: Persuasive techniques in contemporary cults, in Cults and New Religious Movements. Edited by Galanter M. Washington, DC, American Psychiatric Press, 1989 58. Jackson JM: Social impact theory, in Theories of Group Behavior. Edited by Goethals GR, Mullen B. New York, Springer-Verlag, 1987, pp 111–124 59. Perris C: Towards an integrating theory of depression focusing on the concept of vulnerability. Integr Psychiatry 5:27–32, 1987 60. Beck AT: Cognitive Therapy and the Emotional Disorders. New York, International Universities Press, 1976 61. Ruotolo A: Dynamics of sudden murder. Am J Psychoanal 29:162–176, 1968 62. Malmquist CP, Meehl PE: Barabbas: a study in guilt-ridden homicide. Int Rev Psychoanal 5:149–164, 1987 63. Felthous AR: Personal violence, in The American Psychiatric Publishing Textbook of Forensic Psychiatry. Edited by Simon RI, Gold LH. Washington, DC, American Psychiatric Publishing, 2004, pp 471–500 64. Stoll BJ, Kliegman RM: Overview of mortality and morbidity, in Nelson Textbook of Pediatrics, 17th Edition. Edited by Behrman RE, Kliegman RM, Jenson HB. Philadelphia, PA, Saunders, 2004, pp 519–523

280


65. Wissow LS: Infanticide. N Engl J Med 339:1239–1241, 1998 66. Overpeck MD, Brenner RA, Trumble AC, et al: Risk factors for infant homicide in the United States. N Engl J Med 339:1211–1216, 1998 67. Resnick PJ: Murder of the newborn: a psychiatric review of filicide. Am J Psychiatry 126:58–63, 1970 68. Bourget D, Bradford JM: Homicidal parents. Can J Psychiatry 35:233–238, 1990 69. Sichel D: Neurohumoral aspects of postpartum depression and psychosis, in Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Edited by Spinelli MG. Washington, DC, American Psychiatric Publishing, 2003, pp 61–79 70. Kohl C: Postpartum psychoses: closer to schizophrenia or the affective spectrum. Curr Opin Psychiatry 17:87–90, 2004 71. Chaudron LH, Pies RW: The relationship between postpartum psychosis and bipolar disorder: a review. J Clin Psychiatry 64:1284–1292, 2003 72. Lewis CF, Bunce SC: Filicidal mothers and the impact of psychosis on maternal filicide. J Am Acad Psychiatry Law 31:459–470, 2003 73. Hogg W: Mothers who murder their children: an impressionistic study. Am J Forensic Psychiatry 25:45–54, 2004 74. Spinelli MG: Neonaticide, in Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Edited by Spinelli MG. Washington, DC, American Psychiatric Publishing, 2003, pp 105–118 75. O’Malley S: “Are you there alone?” The Unspeakable Crime of Andrea Yates. New York, Simon & Schuster, 2004 76. American College of Forensic Psychiatry: A cautionary note for forensic psychiatrists: Andrea Pia Yates v the State of Texas. Am J Forensic Psychiatry 26:5–14, 2005 77. Pine F: On the origin and evolution of a species of hate: a clinical-literary excursion, in The Birth of Hatred. Edited by Akhtar S, Kramer S, Parens H. Northvale, NJ, Jason Aronson, 1994, pp 105–132 78. Ingebretsen EJ: At Stake: Monsters and the Rhetoric of Fear in Public Culture. Chicago, IL, University of Chicago Press, 2001 79. Macfarlane J: Criminal defense in cases of infanticide and neonaticide, in Infanticide: Psychosocial and Legal Perspectives on Mothers Who Kill. Edited by Spinelli MG. Washington, DC, American Psychiatric Publishing, 2003, pp 133–166 80. Cartwright R: Sleepwalking violence: a sleep disorder, a legal dilemma, and a psychological challenge. Am J Psychiatry 161:1149–1158, 2004 81. Barton B: When murdering hands rock the cradle: an overview of America’s incoherent treatment of infanticidal mothers. SMU Law Rev 51:591– 619, 1998 82. Eliott C, Chambers T (eds): Prozac as a Way of Life. Chapel Hill, University of North Carolina Press, 2004 83. Jick H, Kaye JA, Jick SS: Antidepressants and the risk of suicidal behaviors. JAMA 292:338–343, 2004


281

84. Healy D: Let Them Eat Prozac: The Unhealthy Relationship Between the Pharmaceutical Industry and Depression. New York, New York University Press, 2004 85. Lapierre YD: Suicidality with selective serotonin reuptake inhibitors: valid claim? J Psychiatry Neurosci 28:340–347, 2003 86. Healy D, Whitaker C: Antidepressants and suicide: risk-benefit conundrums. J Psychiatry Neurosci 28:331–337, 2003 87. Khan A, Khan S, Kolt R, et al: Suicide rates in clinical trials of SSRIs, other antidepressants, and placebo: analysis of FED reports. Am J Psychiatry 160:790–792, 2003 88. Healy D: Lines of evidence on the risks of suicide with selective serotonin reuptake inhibitors. Psychother Psychosom 72:71–79, 2003 89. Birger M, Swartz M, Cohen D, et al: Aggression: the testosterone-serotonin link. Isr Med Assoc J 5:653–658, 2003 90. Krakowski M: Violence and serotonin: influence of impulse control, affect regulation, and social functioning. J Neuropsychiatry Clin Neurosci 15: 294–305, 2003 91. van Heeringen K: The neurobiology of suicide and suicidality. Can J Psychiatry 48:292–300, 2003 92. Cherek DR, Lane SD, Pietras CJ, et al: Effects of chronic paroxetine administration on measures of aggressive and impulsive responses of adult males with a history of conduct disorder. Psychopharmacology (Berl) 159: 266–274, 2002 93. Gottlieb R: A psychoanalytic hypothesis concerning the therapeutic action of SSRI medications. J Am Psychoanal Assoc 50:969–971, 2002 94. Levine FJ: On the therapeutic action of SSRI medications. J Am Psychoanal Assoc 52:492–493, 2004 95. Brent DA, Birmaher B: British warnings on SSRIs questioned. J Am Acad Child Adolesc Psychiatry 43:379–380, 2004 96. Satel S: Antidepressants: two countries, two views. The New York Times, May 25, 2004, p D6 97. Harris G: New York state official sues drug maker over test data. The New York Times, June 4, 2004, p A1 98. Rejection of “Prozac defense” in murder trial viewed as reassuring. Clinical Psychiatry News 19:13, 2001 99. Lande RG: Whole blood serotonin levels among pretrial murder defendants. J Psychiatry Law 31:287–303, 2003 100. Motus v Pfizer, Inc, 358 F3d 659 (2004) 101. Miller v Pfizer, Inc, 356 F3d 1326 (2004) 102. Tobin v SmithKline Beecham Pharmaceuticals, 164 F Supp2d 278 (2001) 103. Donovan S, Clayton A, Beeharry M, et al: Deliberate self-harm and antidepressant drugs: investigation of a possible link. Br J Psychiatry 177:551– 556, 2001 104. Whitehead PD: Causality and collateral estoppel: process and content of recent SSRI litigation. J Am Acad Psychiatry Law 31:377–382, 2003

282


105. Coburn v SmithKline Beecham Corp, 174 F Supp2d 1235 (2001) 106. Grady D, Harris G: Overprescribing prompted warning on antidepressants. The New York Times, March 24, 2004, p A13 107. Meier B: Boy’s murder case entangled in fight over antidepressants. The New York Times, August 23, 2004, p A1

CHAPTER

9 JUVENILES AND HOMICIDE

ONE WAY TO VIEW the problem of homicides committed by juveniles is to see it as paradigmatic for the wider problem of serious forms of juvenile violence. The justification for such a paradigm comes from the realization that in quasi-legalistic classifications, such as those applied in the juvenile justice system, an enormous overlap exists in the behaviors of the perpetrators of homicidal types of violence. Cases of attempted murder and the endless cases of aggravated assaults committed by juveniles are separated by a thin line, especially when the violence occurs in group settings. For example, an act of juvenile vandalism in which a baseball bat is swung, hitting another youth in the back of the head, may in one case result in a death whereas in another case only a brief state of unconsciousness. However, the legal consequences for juveniles in such circumstances are far greater than for adults. The consequences can vary from probation in a juvenile justice system without any criminal sanctions on one hand, to certification to an adult criminal court on the other, with all the vagaries inherent in that process—including the possibility of a later conviction for first-degree murder and being sentenced to an adult prison. In this chapter, I discuss these types of cases in the context of some of the psychiatric aspects of juvenile violence that can result in a homicide. Juveniles’ behavioral examples do not make the problem of juvenile homicide per se seem any different than homicides in other age groups. Many of the same types of psychological and social factors that feed

283

284


into homicidal proneness in an adult population operate in juveniles. However, the added dimension in assessing problems of juvenile psychopathology is that these juveniles are not fully formed biological organisms, at least to some degree, and do not have psychological attributes that are solidified in terms of an ultimate character structure. Another dimension is that adults do not like to think of children as being able to commit the same types of homicidal acts as adults. Such a revelation makes the subject of juvenile homicides one of great fascination for the public and the media, who never seem to tire of confessional stories involving adolescents. In all of these cases, the element of a horrific fascination permeates the description of extreme violence perpetrated by an adolescent. Even more horrific to the public are homicidal acts committed by preadolescents against family members or peers. Developmental components loom large in this subgroup. It is important to keep in mind that juveniles who commit homicides do not form a homogeneous group any more than do those who commit specific delinquent acts.1 Even more disturbing are the confusing overlap and lack of clarification between delinquent behavior in juveniles and the possible psychopathology that may be present as well.2 Although it is sufficiently difficult to classify homicides legally, it is even more difficult to classify them by psychopathological states, especially when the homicides are perpetrated by adolescents. With these caveats in mind, I examine various kinds of homicides perpetrated by juveniles. One possibility is that homicide is an individual act in which certain types of psychopathology may be detected and an attendant diagnosis given that partially explains it. Sometimes the act is a homicide perpetrated against one family member, or it may be an act of familicide perpetrated against an entire family. In yet other situations, the act erupts in the context of a violent confrontation between two juveniles, in which one is seriously injured and the other may be murdered. Another alternative is homicide perpetrated by juveniles in groups or gangs. These occur in the context of group dynamics operating on the basis of challenge, provocation, and the need to defend one’s turf or ego. Whatever the setting, by the 1980s, gunshot wounds had become the second leading cause of death among teenagers.3 The problem of juvenile violence itself has been subject to a great deal of controversy. Such violence occupies an enormous amount of time and effort in agencies that deal with juveniles and the juvenile justice system. Unfortunately, most explanations for juvenile violence lack sufficient specificity to confirm what has transpired when a homicide results. Discussions must rely to a good extent on hypotheses or inferences from clinical material and the social context.

Juveniles and Homicide

285

EPIDEMIOLOGY OF HOMICIDES COMMITTED BY JUVENILES The variety of homicidal violence encountered in adolescents is almost legion. The following cases illustrate typical varieties of cases that are processed through juvenile court systems in U.S. cities: • A 16-year-old boy used an ax to kill his entire family. • A 15-year-old boy raped a young woman in her 20s, then panicked, and from fear of discovery, decided that he should kill her. • An adolescent boy robbed a delicatessen. When the store clerk seemed to challenge him, he shot and killed the clerk and ran from the store without taking anything. • An adolescent girl was lured into prostitution. She later assisted her pimp in holding down another girl so that the pimp could cut her throat as a way of disciplining her for running away. The girl died from the throat slashing. • Two adolescents physically pummeled a 35-year-old homosexual man in an episode of gay bashing and then bragged about it in school to their peer group. The parents found it difficult to take the act seriously because they felt “gays were only getting what was coming to them.” • An adolescent boy lived with his divorced father, who was gone most of the week because of his work. The boy invited another adolescent to participate in a plot to shoot and kill his father on his arrival home. The peer first agreed but then reneged. After the father was shot, the friend informed the police. • One night while drinking beer, a group of six adolescents stated their intention to find some people on the streets, whoever they might be, and beat them up. One of the men they beat died. The juveniles and their lawyers made a plea to remain in the juvenile system, rather than being certified to adult criminal court, on the basis that the juveniles had had no intention of killing anyone, and that there had been no premeditation because they simply went out for “kicks” and to “live it up.” • A 17-year-old girl and her girlfriend would periodically go on forays or rob people for jewelry or clothes. One night a young woman was spotted wearing an expensive leather jacket. When she refused to give it up, they shoved her to the ground and shot her. • Two rival gangs challenged each other, resulting in drive-by shootings and some of the members being killed. The subject of violent crime among juveniles in the United States is such an emotionally laden topic that it is difficult to discuss it objectively, even for professionals. Juvenile violence rises and falls parallel to

286


whatever variations are occurring in the area of adult homicidal violence. Two facts must be recognized: violence in the United States is higher than in other Western countries, and serious juvenile violence follows the same patterns as those witnessed in adult violence, although there are differences among ethnic groups. Some of the epidemiological data on juvenile homicides are described in Chapter 1 (“Epidemiological Aspects of Homicide”). A disproportionate amount of crime exists among juvenile males, particularly those in the youth group of 18–25 years; crime also occurs more frequently in juveniles ages 16–17 years compared with younger adolescents. Yet the statistics do not indicate any significant bulge in terms of homicidal violence for juveniles that has not been paralleled by similar increases occurring in adult violent offenses during the same time period. In the past, the rates for extreme acts of juvenile violence simply followed the rates for violence in adults. However, the rates have been changing. For example, in 1986, adolescents under age 18 years constituted 27.7% of the U.S. population, but only 4.1% of all juveniles arrested were charged with violent crimes.4 When a more noticeable increase in overall juvenile violence occurred, it was more likely to be related to an increased number of juveniles in the population at that time, which then was reflected in the overall number of homicides committed by juveniles. Contrasts and correlates can be found with homicide and suicide in the juvenile population. In the age group of 7–17 years, homicide is the second leading cause of death and suicide the fourth (unintentional injuries, such as accidents, are first and neoplasms third).5 In the time period of 1981–1998, 27,000 juveniles died by homicide and 20,775 by suicide. In the 13- to 14-year age group, the numbers of those who committed suicide equaled the number murdered. In contrast to murder trends, which increased for juveniles into the early 1990s, the suicide rate was constant for this age group in the 1990s after peaking in 1988 at 3.4 per 100,000. Although the juvenile suicide rate overall fell 11% from 1994–1998, it was still 28% above that in 1981; for black youths it was 126% higher. However, Asian and black youths were more likely to be murdered than to commit suicide. In the period of 1981–1998, Asian youths were 1.5 times and black youths 8 times more likely to be murdered than to commit suicide. Data from the Centers for Disease Control’s National Center for Health Statistics revealed that by 1991, when homicide rates were peaking, nearly half of the 26,513 homicide victims in the United States were from the age group between 15 and 34 years.6 In that age group, homicide had become the second leading cause of death for males. From 1985 to 1991, the annual crude homicide rate increased 25%, but the rate


287

for persons ages 15–34 years increased by 50%. From 1963 through 1985, the annual homicide rates for 15- to 19-year-old males were one-third to one-half the rates for the next-higher 5-year age groups (20–24; 25–29; 30–34). However, by 1985–1991, the annual rates for males ages 15–19 years had increased 154%, surpassing the rate increase in the next three age groups (which increased by 76%, 32%, and 16%, respectively). During the 1985–1991 period, age-specific arrest rates for murder and nonnegligent manslaughter increased 127% for 15- to 19-year-old males, again higher than the next three age groups. These types of changes in the annual homicide rates for 15- to 19year-old males from 1985 to 1991 reflect a dramatic shift from the earlier period of 1963–1984. Firearm-related homicides accounted for 88% of all homicides in the 15- to 19-year-old male group in 1991 and 97% of the rate increase from 1985 to 1991. Although attaining causal specificity is difficult, it can at least be hypothesized that the increased recruitment of juveniles into the drug market and the diffusion of guns to young people were significant variables. Although part of this pattern is related to gang violence in major metropolitan areas, the phenomenon of gangs in major metropolitan areas is not new in American society. Yet it must also be kept in mind that juvenile gang homicides increased from less than 2% of all homicides in 1987 to almost 5% of all homicides in 1993; in 1987, 317 people were killed in juvenile gang homicides, and in 1993, 1,147 were victims of juvenile gangs.7 What has occurred since then? The numbers of juveniles arrested for murder decreased from 1994 to 2000. In 2000, 1,610 juveniles were murdered in the United States compared with the peak of 2,880 in 1993. Adjusting for the size of the juvenile population, the murder rate of 2.3 per 100,000 in 2000 was lower than that in the preceding 20 years.8 From 1984–1992, the disturbing increase in the juvenile murder rate was followed by the dramatic drop from 1993 to 2000. During the period of 1980–1993, the juvenile male homicide rate increased 117%, and the female rate increased 23%. The increase in the murder rate for black juveniles of 92% was significantly greater than the 20% for white males. By 2000, there were decreases of 50% for males and 38% for females in the juvenile population. Caution must be exercised in interpreting these figures about juveniles because the figures may be skewed by various factors. For example, some homicides in the statistics may be cases in which multiple juveniles acting in a group perpetrated one homicide. The result is then a distortion in that they would have all been legally arrested on some level and charged with a homicide as a group offense to clear one homicide case. In fact, more group killings of this type are committed by ju-

288


veniles and youths than adults. At the time of a murder, public fury often increases demands for an arrest, resulting in some overcharging. The charges made at the time of the arrest may subsequently be lowered in the course of plea bargaining and delays in the case, a favorite legal defensive maneuver. Ultimately, some of the charges may be completely dismissed. Among all victims of violent crimes involving juvenile offenders, 17% involved multiple juveniles acting together and another 15% involved juveniles and adults acting together, making almost a third involving multiple juveniles.9 The 1990s thus saw an unprecedented epidemic in youth violence.10 The violence was demographically concentrated among black male youths, among whom the homicide commission rate increased by a factor of 4.5. One of 10 homicides was committed by a juvenile younger than age 18. A decline in the adolescent population was offset by an increase in the rates of arrest. The data indicated more youths killing than being killed, but with a crossover in both directions. Every category of homicide exhibited an increase in gun use. Some criminologists raised the issue of whether juveniles had become more vicious. The emphasis was on the defects in socialization processes that contributed to impulsivity and homicidal juveniles. The defects noted were lack of supervision by parents or other adults, the prevalence of illegitimacy, singleparent households, and television as the predominant socializing influence, with its portrayal of violence as the solution to problems. For some juveniles, the influence of the drug trade turned gangs and guns into the symbols of respect, identity, power, and manhood.11

Nature and Background of Crimes Committed by Juveniles Some researchers argue that violent offenses ascribed to juveniles might not actually involve the occurrence of significant physical harm to a victim during the commission of the offenses. For example, results from one study from Columbus, OH, involving 811 juveniles with at least one violent crime in their records revealed that 72% of those juveniles did not threaten or inflict significant physical harm during the commission of the violent offenses.12 Authors of another study from 1981, in which a cohort of 50,000 youths were followed from birth into adulthood, found that only 22 of this group later committed two or more aggravated offenses in which physical harm was inflicted or even threatened.13 However, the significant variable may not be whether physical harm occurred but rather the occurrence of violent crime itself. It does


289

appear that serious violence occurring at a young age is a premonitory sign of later violence. Hamparian and colleagues14 have shown a continuity between violence committed by a juvenile and violence committed by an adult. Although there is a paucity of adequate data about violent crimes and criminal careers, in one study of a London cohort it was found that the majority of juvenile violent offenders went on to commit adult nonviolent offenses. In that study, the authors did not find that the majority of adult violent offenders had been previously convicted of violence as a juvenile.15 The role of some type of antecedent physical and/or sexual abuse for juveniles who commit a homicide raises many questions. The related problem of the impact of witnessing parental violence, which may not show up immediately in the child, also needs consideration. There is also the problem of more massive exposure to violence in the media itself. Although reports of child abuse increased from 60,000 in 1973 to 3,000,000 in 1993, it is not clear whether this number reflects an actual increase in child abuse or rather an increase in reporting.16 Longitudinal studies have compared abused and neglected juveniles with those without such a background. For abused and neglected juveniles, these studies show an increased risk of juvenile delinquency or later adult criminality and a diagnosis of antisocial personality disorder.17 When child maltreatment is assessed in a public school setting, similar findings show an increase in violent and serious delinquency in the maltreated children.18 However, all these studies do not establish an increased risk for homicide per se. Although a logical inference may be that homicidal juveniles will emerge from this group of abused, neglected, and maltreated children, the distinction between serious delinquent behavior and homicidal behavior is still relevant. This distinction is discussed later in more detail (see sections in this chapter, “Difficulties in Diagnosing Juvenile Violent Behaviors” and “Role of Childhood Abuse”).

Weapons and Juvenile Crime Pertinent to this discussion is the recurring question of whether juveniles usually use weapons in committing violent offenses. In one early study, Wolfgang and colleagues19 followed all the boys born in Philadelphia, PA, in 1945 and then those living in the city between their 10th and 18th birthdays; they found that of the 9,934 offenses known to police, weapons were involved in only 263. In a 1970s study in New York City, weapons were found to be involved in only 17% of violent juvenile offenses.20 However, data taken from victim crime surveys have indicated that the

290


use of weapons was higher, with juvenile offenders using weapons in 27% of the offenses, although guns were used in less than 5% of the cases.21 The current significance of firearms connected to murders in the adolescent age group is striking. In 1990, gunshots caused one in four deaths in the adolescent age group. In 1990, 4,200 teenagers were killed by bullets, compared with 2,500 deaths only 5 years earlier.22 Almost half of the black males ages 15–19 years who died in the United States in 1989 were murdered, usually with a gun.23 In 1989, homicide victims ages 10– 19 years numbered 2,771, with 80% being shot with guns and 10% stabbed. This rate was noted to be a 140% increase from 1979 figures for black males in that age group.24 The major victims of juvenile violence are other juvenile peers, except for the victims of purse snatchings. Much of these results are attributed to the increase in gun carrying among youths. National data have revealed that among high school students, the monthly prevalence of weapon carrying in 1990 was 20%, with the highest incidence found among black and Hispanic males (39% and 41%, respectively).25 However, little is known about the variables responsible for this increase. One study of two public junior high schools in Washington, DC, found that although knife carrying was associated with aggressiveness, it was not related to serious delinquency; however, gun carrying did appear to be a component of highly aggressive delinquency and not just a purely defensive behavior.26 In Seattle, WA, a survey of half of the public high schools indicated that 34% of the students reported having easy access to handguns, with the highest prevalence reported by black male students at 59%.27 The debate on gun control goes on. It has been noted that the epidemic of youth homicide was gun driven, because non-gun homicide rates remained essentially unchanged. This explanation has often been connected to an increase in the prevalence of crack cocaine use by youths, who then acquire guns within the drug subculture. However, gun killings increased in other contexts as well, such as in domestic homicides. When the homicide rates began to decline, they declined for gun as well as non-gun homicides. At the peak of the homicide epidemic in 1993, the percentage of homicide victimization by guns was 90% for males ages 13–17 and 88% for ages 18–24. However, even after 4 years of decline in the homicide rate, by 1997 the percentages for gun use had only dropped one point.28 What effect legislation will have that allows the carrying of concealed weapons is problematic at this time. Those who support this measure believe that it will lower the risk of becoming a crime victim, through defensive gun use (“if guns are outlawed, then only outlaws will have guns”). Opponents argue that gun availability increases the potential for guns’ lethal use.


291

PERSONALITY DISORDERS One of the more controversial topics at present is the relationship of later adult disturbances to personality disturbances that originate in childhood and adolescence. These early diagnoses are important with respect to homicides among juveniles because whatever later diagnoses are attached to those who commit a murder, the possible developmental relationship of the two diagnoses to a homicidal outcome must be examined. These interrelationships can be quite complex, because what may promote violence at one stage may not be as significant at another. A developmental perspective supports a viewpoint that those who end up as chronic or persistent offenders began deviant behavior as children. Loeber and Farrington29 did a review of criminal behavior and found that the youngest criminals have a long history of disruptive behavior beginning in early childhood with truancy, lying, theft, and cruelty to animals. Yet although in the period 1980–2000, 600 children age 12 years and under committed a murder, the specificity question persists about whether these developmental components predispose toward homicidal-level violence in contrast to an outcome of more general types of delinquency and criminality. A question can be raised of whether psychiatric diagnoses would provide more specificity to solving the development toward later homicide. There are always persistent questions in trying to assess the role of personality factors and psychopathology in juveniles who commit homicides. Such questions point out differences in conceptualizing the problem as well. Even apart from legal classifications based on such factors as intent, provocation, self-defense, age of the perpetrator, and duress, differences exist in the approaches used in the behavioral and social sciences. As noted, a developmental approach looks for significant antecedents that may contribute to such a final outcome as a homicide. Some seek to use an empirical approach of having a control group compared with homicidal juveniles. The difficulty in using such a gold standard in homicide cases is that someone in the control group may have committed homicidal-level acts. One study used a juvenile court population to compare 71 juveniles who had committed homicides with 71 nonviolent delinquents who had a heterogeneous mixture of offenses. The former had an increased incidence of criminally violent families, more educational deficits, participation in gangs, and alcohol abuse.30 Using data from subjects’ psychiatric evaluations, a retrospective study covering 10 years compared 39 juvenile males charged with murder with 62 juvenile males charged with some other violent offenses.31 Interestingly, although 33 variables

292


were studied, only three factors were found significant. One was that the group charged with murder was less often given an Axis I psychiatric diagnosis, and half of the homicide group was simply diagnosed with an adjustment disorder or substance abuse disorder. A second finding was that the homicide group more often acted alone, with 46.7% having done so compared with 8.1% for the nonhomicidal group. The third significant finding was that the homicides occurred in a domestic setting in 40% of the cases compared with only 6.5% for the other offense group. An English study compared 40 juvenile murderers with 106 juvenile arsonists over a 10-year period.32 Although the groups were matched for demographic data such as ethnic background, socioeconomic level, and criminal history, the question arises about what substantive knowledge is gained in a comparison with a group of fire setters. The significant differences between the groups were that the homicide group members were more likely to be male, to abuse alcohol, to have been intoxicated at the time, and to have frequently changed schools. They were also less likely to be diagnosed as psychotic. A discriminant function analysis found that only alcohol abuse at the time of the offense in the homicide group was more prevalent.

Cautions About Applying Personality Disorder Diagnoses to Children Clinical descriptions are customarily based on direct contact with or evaluation of the juveniles who have committed or are charged with a homicide, although some may be based on chart or file research. Within this approach, theoretical orientations vary. Some are seeking personality attributes, whereas others look for descriptive symptoms that go with a DSM diagnosis. Within the clinical framework, some psychodynamic explanations may be offered as to why a murder has occurred. Heide 33 summarized some of the personality characteristics she found in adolescent homicide offenders. She found them lacking in selfesteem, with deficits in communication skills and decision-making abilities. Although some appeared tough and cool, deep down they were seen as insecure and lacking a belief that they could succeed in activities such as school, work, or sports. They also showed deficits in dealing with strong negative emotions, such as anger or jealousy, which often led to poor judgment. A low frustration tolerance and a compulsion to strike back were also observed. Their tendency to misinterpret provocations seems parallel to those discussed with borderline problems (see Chapter 4, “Borderline Personality Disorder and Homicide”). Heide’s observation was that juvenile murderers seemed more bored than an-


293

gry, in the sense of not being committed to conventional or prosocial activities. Engaging in antisocial behavior seemed like a way to reduce their boredom. This description would raise questions about the presence of some type of affective disorder. In DSM-IV–TR,34 children are not recognized as having distinct personality disorders. The result is an assumption that disorders in the development of their personalities are contained within the grouping of “attention-deficit and disruptive behavior disorders,” which includes attention-deficit/hyperactivity disorder (ADHD; both specific types and not otherwise specified), conduct disorder, oppositional defiant disorder (ODD), and disruptive behavior disorders not otherwise specified. The reasoning of those who believe personality disorders should not be diagnosed in children is that the thrust of development, and its fluctuations, is such that children’s patterns or traits are not yet stable and may later shift. The position is that such factors make it difficult to diagnose personality disorders in childhood and that therefore it is more desirable to use other diagnoses. Another objection to labeling children with personality disorder diagnoses is the social stigma attached to the diagnoses as well as the tendency that once a diagnosis is applied to a child, that child will have that diagnosis in perpetuity. The hope remains that intrapsychic alignments or the forces of development will resolve whatever is promoting certain kinds of antisocial behavior. If developmental phenomena are such that evolving patterns do not tend to overcome the types of chronic distortions in development that are occurring in the child’s personality, then one of the childhood diagnostic categories is supposed to function as an interval diagnosis. Subsequently, those personality traits still persisting once a child reaches age 18 years can be given the corresponding adult personality disorder diagnosis. If they do not persist, then not having labeled the person with the diagnosis in childhood is seen as having achieved a desirable outcome.

Juvenile Behavior Correlating With Antisocial and Borderline Personality Disorders An initial line of inquiry is whether some predictive trends can be seen in childhood behavior patterns that later become relevant to homicidal types of violence in adolescence. Part of the answer to this inquiry lies with future clinical work and research; through these, greater validity in terms of correlating various childhood diagnoses with those found in adulthood may be attained. A congruency between childhood symptoms and adult symptoms, or greater clinical ability to trace the unfold-

294


ing symptoms and states or traits corresponding to adult personality disorders, would also allow a better prediction of future diagnoses and perhaps of violence. Thus, in terms of the three-clusters approach used in DSM-IV-TR, persistent shyness in childhood or adolescence may correspond to a diagnosis of avoidant personality disorder in adults in Cluster C; ODD may correlate with several personality disorders in Cluster B, and certain developmental disturbances may correspond to schizotypal personality disorders in Cluster A where there are introversion and unusual cognitions. In turn, the various manifestations of conduct disorder could be viewed as reflecting some type of personality disorder cluster that eventually will end up in the borderline, narcissistic, or antisocial personality disorder grouping. The developmental model in DSM-IV-TR is that disruptive behavior disorders are not transient but stable disorders and that ODD can be a precursor to conduct disorder, which in turn can be a precursor to antisocial personality disorder. Yet only a proportion of children with ODD develop conduct disorder, and only a proportion of those with conduct disorder later meet criteria for antisocial personality disorder. To complicate matters, ODD and conduct disorder frequently co-occur with other psychiatric conditions.35,36

Antisocial Behavior Currently, the criteria for the diagnosis of antisocial personality disorder require that a conduct disorder be present before age 15 years and that antisocial personality disorder not be diagnosed until age 18 years. However, it is likely that in the future, these restrictions will be removed and diagnoses of antisocial personality disorder as well as borderline personality disorder will be allowed during adolescence. The reasons are that more astute diagnoses and research should reveal manifestations of these disturbances as persistent over time, and their salient features or forerunners are emerging quite early in personality development. Kernberg37 has stated her belief that a diagnosis of antisocial personality disorder should be made from age 15 years on when a lack of guilt and remorse is present and at least four of the following patterns are found: • An inability to sustain consistent work behavior or the equivalent behavior if the person is a student • Failure to conform to social norms of lawful behavior • Irritability and aggressiveness as indicated by repeated physical fights or assaults • Failure to honor financial obligations

Juveniles and Homicide • • • •

295

Failure to plan ahead, or impulsivity Lack of regard for the truth Recklessness with regard to personal safety or the safety of others Lack of remorse or feeling justified in having hurt, mistreated, or stolen from another

If the diagnoses of antisocial personality disorder and borderline personality disorder were allowed in adolescents, they would be quite frequently found in individuals who commit a homicide. Apart from a few blatantly psychotic juveniles, most homicidal acts by juveniles occur in those for whom an earlier diagnosis of conduct disorder in childhood could be made. As adolescence continues, these individuals’ behaviors appear to be some precursor of either a borderline personality disorder or an antisocial personality disorder. Occasionally, a juvenile with more striking paranoid symptoms seems to express this thinking in the context of a fragmented type of borderline disorder organization or schizophreniform behaviors. A combination that may be seen in those with externalizing disorders, such as conduct disorder and ADHD, is the added presence of alcohol and drug abuse, often comorbid with Cluster B personality disorders.38 This mixture is another found in juveniles who have committed homicides. Unfortunately, the younger child who presents symptoms of some variation of a disruptive disorder is simply fitting into a descriptive appellation that includes behaviors that can vary from homicidal violence to stealing, truancy, absenting from home, and arson. To make matters more nebulous, four main groupings of conduct disorder are listed in DSM-IV-TR: 1. 2. 3. 4.

Aggression toward people and animals Destruction of property Deceitfulness or theft Serious violation of rules

Whether these categories will achieve any greater validity or reliability than earlier classifications of group type, solitary aggressive type, or undifferentiated type is problematic.39 In one 4-year follow-up study, Cantwell and Baker40 found that the diagnosis of conduct disorder was not stable and that after 4 years, many of the initial diagnoses of conduct disorder had been changed to other diagnoses. These diagnoses may have changed to milder diagnoses, such as adjustment reaction, whereas in other cases the behaviors may have progressed to a more prominent type of antisocial behavior similar to adult personality

296


disturbances. However, in addition to the disruptive disorders, some type of mood disorder may have surfaced that could be related to aggressive or violent behaviors.41 Some researchers have suggested that a neurological central nervous system dysfunction, along with learning disabilities, could contribute to an individual’s tendency toward aggressiveness. These factors are all seen as potentially predisposing one to impulsivity or violent behavior. Some work indicated that biochemical measures may present another dimension in assessments of aggressivity, measures that would overlap with various diagnoses along a neuropsychiatric continuum. One group found a correlation between low levels of 5-hydroxyindoleacetic acid (5-HIAA) in the cerebrospinal fluid and the severity of physical aggression in a 2-year follow-up of children and adolescents.42 Another group that assessed central serotonergic function in aggressive and nonaggressive boys with ADHD by way of prolactin response to fenfluramine challenge found that the aggressive group had a significantly higher response.43 This finding correlates with suggestions that serotonergic sensitivity in those with borderline personality disorders contributes to the individuals’ fear and anger and thus their impulsiveaggressive behavior.44 Impulsivity is seen as having a relationship to decreased central serotonin activity as measured in cerebrospinal fluid and platelets.45

Borderline Behavior The role of borderline behavior in adolescents who are involved in homicides is a recurring issue. Some of their instability can produce, just as in adults, impulsive outbursts connected with rage reactions. Until recently, it was customary to diagnose adolescents who appeared to be showing the onset of borderline personality traits as having an identity disorder; now the trend seems to be to extend thinking about borderline personalities to younger ages. The clinical picture portrayed is a mixture of both internalized conflict and externalized behavior. Although adolescents with borderline behavior may be depressed and suicidal and even show micropsychotic symptoms, they may be impulsive. Rapoport and Ismond46 advocated that children and adolescents be diagnosed with borderline personality disorder, provided that the criteria for that disorder were met and the nature of the disturbance was pervasive, persistent, and not limited to a developmental stage. Children develop a sense of identity at quite young ages, but those who appear headed for a fragmented development are plagued by problems in this area. In these troubled adolescents, lower levels of defensive operations begin with identity


297

diffusion, which is seen as part of a personality disorder disturbance. Borderline defense mechanisms in the form of denial, projection, primitive idealization, splitting, devaluation, and omnipotent control then appear. The relationship of the juvenile to reality becomes more tenuous, although the capacity to test reality is preserved. In some of these juveniles, a much more severe type of disorganization may occur that appears to be similar to a psychotic breakdown. It is for this reason that some believe that borderline pathology of childhood should not be viewed as simply an earlier version of the adult diagnosis. The descriptive construct of “multiple complex developmental disorder” is seen as more valid because it has more implications for serious longterm sequelae.38 A fragmented and bizarre sense of identity may emerge in which the individual begins to have a severe constriction in personality functioning and to deanimate animate relations, so that he or she begins to react to people as though they are robotized or simply objects. Animation of inanimate or nonhuman objects is also seen in the way these juveniles react to a toy as though it had some lifelike qualities or to a pet animal as though it were someone who could communicate with them on their own level. These behaviors appear to be last-ditch efforts to stem the occurrence of overwhelming anxiety or depression; by then there are cognitive distortions with a loss of the capacity to test as well as relate to reality. In some homicides, these areas of psychopathological development are reflected in the juvenile perpetrator’s numbing response to the dead victim, as though the victim were an inanimate object. During the homicidal act, dissociation and depersonalization seem to permit the act to occur. In discussing psychotic behavior in juveniles, Kernberg37 cited the following case example of murderous behavior.

Case Example Maria, a 15-year-old girl, displayed borderline behavior. She tried to kill her parents by cooking stews and muffins mixed with rat poison; she also set fires outside their bedroom. Maria referred to her parents, who had come from Latin America, as “minority WASPs.” Except for one friend, she felt no one could be counted on. Her parents, who supposedly ignored her sensitivity by the way they ate and talked, disgusted her. Furthermore, she felt that she could never forgive her parents for not controlling an older brother who had physically abused her.

When adolescents with a borderline personality are examined after the commission of some kind of homicidal act, their personal history usually reveals that certain developmental landmarks have not been attained. Characteristically seen are a history of unresolved massive sep-

298


aration anxieties on entering school or going away to a camp; a lack of consistent internalized standards in terms of what is acceptable or not acceptable behavior; often a history of abuse or neglect; a history of impulsive acts occurring with a sense of accompanying unpredictability; and difficulty in modulating the expression of affects in diverse types of situations. By the time adolescence arrives, they are further confronted with the increased confusion that goes along with their developing sexuality and its expression. The problems that can result from the interaction of borderline personality characteristics and developing sexuality are seen in the following case example.

Case Example An adolescent male had been in treatment at a residential treatment center for 1 year. While there, a girl of his own age had promised to have sex with him one night if he came to her room. However, after arriving at her room, he found that she had changed her mind, with the result that he began to utter a string of profanities at her along the line of “bitch,” “slut,” and so on. This tirade prompted a female staff member to come to the girl’s room to inquire about the commotion. When she asked the youth to refrain from such language, he swung his fist at her, knocking her to the floor, and then kicked her in the head. After being restrained by other staff members, he was asked about the incident. He stated, “She had an attitude problem. She had no right to step in because it was between this other girl and me.” He had no sense of the inappropriateness of his behavior in assaulting the staff member; instead, a determined self-righteousness about his behavior prevailed. Thus, this youth’s lack of empathy for others—whether for staff or members of his peer group—was documented before his committing a homicide 1 year later.

The relationship of juveniles with borderline personality qualities to their peer group alternates between feeling intimidated and coerced by their peers on one level and a need to control and dominate them on another level. These juveniles are able to idealize certain adults or older peers but are also liable to devaluate and turn on them at any time. In this context of ambiguity and uncertainty, some type of homicidal behavior can occur. Most striking is these juveniles’ inability to see the role their provocative behavior plays in assaultive behaviors—which may lead to a homicide—as well as their inability to accept responsibility for their own actions. In a psychodynamic sense, the separation-individuation process, along with other developmental components, has become derailed in these juveniles. What occurs instead are attempts at grandiosity intermixed with identity diffusion. Although sufficient explanations are lacking, it appears that the behaviors are a defense against an overwhelming


299

anxiety of feeling alone and abandoned, much as occurs in adults with borderline personality disorder. If narcissistic components are dominant in the disorder, these behaviors appear to be an effort to overcome a sense of worthlessness and shame by conveying an arrogance and devaluing others. In this vulnerable state, provocations are more likely to elicit reactions that can lead to physical violence, if not homicide, when controls are impaired. A deformity in these juveniles’ superego development prohibits the proper modulation of actions. Instead, what emerges are manifestations of extreme ambivalence as well as outright hatred, expressed within their peer group or family. Pervasive types of splitting mechanisms are observed in which groups and people are split into the customary good guys or bad guys. Shifts in the splitting within the group and between certain individuals further contribute to the potential for a violent outburst. When others become puzzled or confused by the results of these changes going on within a juvenile, misinterpretations of the juvenile’s behavior are likely, which may lead to confrontations. This result is particularly true because of the sense of grandiosity and entitlement present in these adolescents, who try either to control others totally or to look for someone to subsume their identity under and become an obedient follower. Hence, causes or groups that these youths can join, including the gangs seen in major metropolitan areas, appeal to them.

Difficulties in Diagnosing Juvenile Violent Behaviors Several difficulties exist in relying on diagnostic groupings such as disruptive behavior disorders to assess homicidality. Although a retrospective diagnostic assessment of juveniles after a homicide may often find a loading from within the disruptive grouping, this extreme degree of violent outcome would be overpredicted by a wide margin. Studies from adult populations have established a relationship between psychopathy and violence.47 In turn, this connection has led to an interest in “juvenile psychopathy” and its possible relationship to serious violence. The existing evidence indicates a moderate association between measures of psychopathy and various forms of aggression. The complicating factor is the developmental changes that occur during adolescence, which make long-term predictions difficult. The difficulty in these predictions is related to the problem of distinguishing between violence risk assessments per se and the validity of the construct of psychopathy in juveniles as it may relate to homicidal levels of violence.48 There is also the recurrent question of antisocial personality disorder in adults being preceded by some type of conduct disorder historically in

300


childhood, but predicting this outcome from childhood is similarly quite unreliable. Some studies have found that conduct disorders predict later criminal behavior.49 However, other studies have indicated that only about one-third of conduct disturbances persist into the adult period.50 The answer to which children will persist and progress in their disruptive behavior remains elusive. Although the possibility exists for a variety of antisocial behaviors (e.g., delinquent acts, criminality of diverse forms, illicit drug use) to evolve, perhaps the most that can be said is that aggressive behavior of some kind is the variable that antedates criminal behavior in 75% of the adult cases.51 One of the pervasive difficulties in studying juveniles who murder is that the overwhelming number of studies focus on “violence,” which is often defined differently. Even welldesigned studies using categories of trying to predict relational violence (persistent hitting) and predatory violence connected to earlier deviant behavior (poor grades, weak elementary school bonds, and a pro-drug middle school environment) leave a big gap before getting to homicidallevel behavior.52 As noted earlier, according to criteria listed in DSM-IV-TR, evidence of a full-fledged conduct disorder before age 15 years is necessary for the diagnosis of an antisocial personality disorder. Thus, a pervasive pattern of disregarding and violating the rights of others must have begun in childhood and early adolescence and continued into adulthood. The personality disorder is seen as unfolding over the years. Note that the DSM-IV-TR criteria for conduct disorder are not actually symptoms but signs. The theoretical basis for this approach is that the assessment of conduct disorder is grounded in signs, although the behaviors can actually be present with diverse diagnoses. The criteria for a particular diagnosis for juveniles, such as conduct disorder, can have similar symptoms appearing with other diagnoses for juveniles. Thus, there is a lack of specificity connected to the diagnosis. Hence, a 16-yearold who grabs a hammer in anger and attacks another person by beating the person’s head and killing him or her would meet the overt criteria of a conduct disorder, now diagnosed as an antisocial personality disorder if some of the other criteria were met. Perhaps one of the difficulties in diagnosis is that DSM cautions against diagnosing personality disorders in adolescents, but both longitudinal and epidemiological research suggest that personality syndromes are recognizable in adolescence.53 However, simply using Axis II personality disorders for diagnosing adolescents also has its difficulties. The prevalence of comorbidity with more than one personality disorder and the categorical approach to the diagnosis (being either present or absent rather than using a dimensional continuum) handicaps the assessment of adolescents.


301

The limitation to the current approach is the lack of specificity and that the approach does not indicate the other possible diagnoses. For example, some type of depressive disorder or predisposing neuropsychiatric impairment might be present that makes such impulsive acts more likely. The possibility of psychotic ideation in which reality testing has been impaired also cannot be automatically dismissed. Another explanation could be that of a fragmentation of the personality, in which the victim attacked has been devalued from a previously idealized state. The process is one in which criminological or legal classifications have slipped into psychiatric descriptions, and diverse violent and disturbed adolescents then get classified as “antisocial.” In assessments of adolescents who have engaged in murderous behavior, the question is raised about the significance of the perpetrator’s facade of indifference or callousness, which is so often present. In many of these adolescents, an underdeveloped system of controls and, for some, an outright distortion in their thinking are present. It is easier for these adolescents to adapt to the world in which they live, particularly if it is a peer group of adolescents or a gang, by putting on a facade of being tough or macho. However, the facade cannot conceal a type of hypersensitivity or the outright distortion or misinterpretation of others’ actions or intentions. Whether this is ex post facto rationalization is a key question. The following case example illustrates this type of mixed sensitivity and vulnerability.

Case Example A gang attempted to burglarize a home. In an effort to conceal their identity so the homeowner could not see them, gang members threw a blanket over him after they knocked him to the floor. The owner’s struggle to get up elicited a rage reaction on the part of one of the juveniles, who proceeded to jump on top of the blanketed figure and pummel him, first with his fists and then with a lamp he grabbed. The owner eventually died. The youth later explained his behavior by saying that the owner should not have struggled in the way he did because he should have realized the gang members’ purpose in throwing a blanket over him was simply to protect themselves from being identified. This youth’s putting the blame for the owner’s death on the owner indicated a type of egocentric thinking in which events were explained in terms of the perpetrator being provoked by the nonconforming acquiescence of the person being burglarized, who, the perpetrator felt, should have been empathetic with the gang’s burglarizing.

Extreme forms of antisocial behavior always carry the potential to culminate in acts of homicidal violence. The question then raised is why these behaviors occur and result in a homicide. A sociological viewpoint sug-

302


gests that a certain group of adolescents behave in antisocial ways as a primary way of relating to others in society. It could then be predicted that for some of them, progressive types of violent acts, including homicide, will occur. Thus the behavior is seen primarily as a cultural or environmental expression of these adolescents’ antisocialness. A contrasting approach seeks some underlying basis to the antisocial behavior, be it on a dimensional or categorical basis. Diagnostic categories are sought that, it is hoped, can be specific enough to explain the behavior. Some explanatory models also seek to go beyond descriptive levels by adding in developmental deviations.

SCHIZOPHRENIC DISORDERS Some juveniles who commit homicide carry the diagnostic signs and symptoms of schizophrenia. Although composing a small number, these juveniles should not be overlooked, because of the clinical and legal ramifications that accompany the diagnosis. On the other hand, it would be erroneous to limit the discussion of all possible psychotic conditions in juveniles to those occurring in conjunction with schizophrenia. It is unfortunate that the criteria used for diagnosing schizophrenia in juveniles vary among studies, which raises questions about the studies’ validity and comparability; as a consequence, data on the incidence of this disorder in juvenile offenders will vary as well. For example, in one study of 14 cases of adolescents condemned to death, several were diagnosed as having a psychotic disorder before their incarceration.54 Benedek and Cornell,55 in their study of 72 adolescents who had committed a homicide, found only 5 (7%) of the cohort were psychotic at the time of the commission of the homicide. Labelle and colleagues56 studied 14 adolescent murderers but attributed a psychotic diagnosis to none of them. Several decades ago, the psychiatrist Loretta Bender57 wrote about “pseudopsychopathic schizophrenia” in which she identified children who she believed were psychotic in childhood but by adolescence had become antisocial and engaged in violent behavior to a homicidal extent. A British study of 39 psychotic adolescents divided them into a “violent” group of 14 who had committed murder, attempted murder, or armed robbery, and 25 who were “nonviolent.”58 Significant differences were found in that the violent group had experienced physical or emotional abuse, engaged in previous criminal behavior, and had contact with social or mental health services. Certainly it is difficult to ignore clinical evidence that some adolescents have an operational delusional system that can induce them to take action on that basis. Similarly, an adolescent reporting hallucinations could respond to hallucinatory commands. Lewis and colleagues59 eval-


303

uated children who had committed a murder and found that based on interpretations of their history, a significant number had previously experienced psychotic symptoms (e.g., hallucinations, illogical thought processes). The most common symptoms associated with these adolescents’ violent acts were paranoid ideation and misperceptions that seemed associated with a brief psychotic state. Perhaps the description of episodic dyscontrol syndromes most closely fits these behaviors in which the psychosis becomes more rampant, or ego controls over violence are lessened to the degree that a homicide occurs. In the absence of some previous confirmatory signs or symptoms of schizophrenia, this syndrome is debatable, but it is often raised in cases in which a seemingly senseless and brutal murder has occurred. The following case examples illustrate acts of parricide occurring in the context of delusional and hallucinatory symptoms in adolescents.

Case Examples A 15-year-old male who had murdered his father faced certification for trial as an adult. At age 8 years, he had been sent to a child guidance clinic for reclusive behavior displayed in his avoiding peer activities. He progressed through school as a loner, having the occasional solo friendship, although these never seemed to last. Within his family of four siblings, he seemed different than the others in terms of his preference to be alone. He progressed satisfactorily in the academic world until he began to be absent from school a good deal during eighth and ninth grades without providing a sufficient explanation. Another referral was made to a clinic, but on review of the case file subsequent to the killing of his father, it was noted that he had only talked about superficial things and not about any deeper feelings or confused thoughts. On the night of the homicide, the youth had been up all night debating within himself whether the pain and suffering he had been experiencing over the past several months was somehow attributable to his father’s influence. The content of his obsessional brooding was that his father had treated him differently than the other children by holding abnormally high expectations for him. During the course of a restless night marked by no sleep and endless pacing, he decided to do something to end his misery. It appears that hallucinations were prominent in terms of his hearing repeated commands to “get it over.” In response to this, he took one of his father’s hunting guns and posted himself at the exit of the garage where he knew that sometime that morning, his father would exit on the way to work. As his father pulled out of the driveway, the youth leveled a shotgun at his father’s head, killing his father. An 18-year-old had been discharged from a residential treatment center where he had resided for 1½ years for treatment of schizophrenia. Over several previous years he had experienced auditory hallucinations to kill one or both of his parents. There had been periodic marijuana use

304


but no other street drugs. On several occasions, he would report his voices to his parents, who would take him to a hospital where he would be admitted for a few days. On one occasion when home alone, he called the police himself, and they took him to a hospital. The discharge from the treatment center occurred because of his reaching age 18, at which time welfare funds for his stay ended. He was discharged to his parents’ home, where he resumed smoking marijuana to an increasing degree along with a return of the old command hallucinations. He awoke one morning hearing commands to kill his parents. He waited for his mother to arrive home from work in the afternoon and shot her with a family gun he obtained by breaking open a storage case. His thinking was to wait for his father to arrive and shoot him as well. However, after shooting his mother, he later described himself as “coming out of the nightmare.” His mother (who later died from her wounds) asked him for a phone and called an emergency number. The boy sat outside with the gun debating whether to suicide when the police arrived.

DEPRESSION Antisocial behavior, including homicidal violence, occurs more frequently in depressed adolescents than is indicated by official statistics. Although depression is routinely associated with suicidal acts, it also contributes to antisocial behavior and may contribute to homicidal behavior just as frequently as it does to suicidal behavior. Much of the discussion involving adult depressions and homicide (see Chapter 8, “The Depressed or Bipolar Person and Homicide”) is relevant to adolescents and to some younger children as well. What appears to occur is a sense of helplessness and doom that progresses to an affective state of needing to resolve a crisis rather than stay in a painful state of hopelessness. One option is suicide, but in what seems a paradoxical solution, homicide appears to offer more hope for alleviating the situation because it will salvage the potential perpetrator’s own life. It is by way of this misplaced type of aggression, perhaps as an act of desperation, that a homicide takes place. In some cases in which a suicide or violent act has been considered, acts such as robbery or arson are carried out instead, which are then disowned by the perpetrator even though the evidence incriminating the perpetrator is overwhelming. As in most of these types of ultimate quests for relief, the act is usually directed toward a loved one, which is seen as an expression of the intense ambivalence that is present in severely depressed people. It should not be forgotten that depression in adolescents, like that in adults, can be of psychotic proportions and affect their behavior through distorted thinking and delusions of hopelessness.


305

In a study conducted through a juvenile court system over several years with an overall population of 6,500 alleged delinquents, there were 213 motions to certify certain juveniles to adult court. The alleged crimes were mainly those of extreme violence, such as a homicide or an aggravated sexual assault, although a few were included on the basis of two or three prior felony arrests.60 By use of criteria from the Minnesota Multiphasic Personality Inventory,61 the Beck Depression Inventory,62 and DSM-III (for a major depressive episode),63 44 of those 213 defendants were diagnosed as being depressed. It is helpful to look at some of the dynamic factors operating in this group in addition to what seemed to be a predisposition for several first-degree family members to have problems with alcoholism and/or antisocial behavior. The various ego defenses used in this group of adolescents did not appear to have been working even before the violent episode because the adolescents had not been able to contain their various emotions and impulses related to aggression. Questions remain as to why this group’s depressive tendencies were expressed in such violent acts rather than through the emergence of some other course of psychopathological development. Although commission of a homicide by an adolescent is not seen as “normal” behavior, a question arises as to when certain signs of impending action should be assessed as clinically significant, such as an adolescent beginning to tell peers that he or she is thinking of killing someone or committing some violent act. Related questions are when such signs should be considered indicative of a major mental disorder, and how to determine in the course of an assessment whether the individual’s defenses will be able to contain the hinted-at aggression. Many of the 44 juvenile males in the study just mentioned had adopted hypermasculine roles but were inhibited in their capacity for any extended gratification through intimate relationships. They resisted close relationships and attempted to conceal their dependency needs. They displayed similarities to those juveniles whose dependency needs are met through gang participation, except in most of the cases the homicides did not occur in the setting of gang violence. For some of the youths, a strong identification with an aloof or absent father left them lonely and angry. When these youths began a progressive withdrawal into their inner world, their behavior initially appeared as a denial of anxiety or agitation. Attempts at forced gaiety would occur in the context of drug or alcohol use for some. In retrospect, a good deal of sadomasochistic behavior with low self-esteem was present in these adolescents, manifested in their provoking others or damaging their own image or status. In some cases, an unwitting shoot-out with the police occurred, with suggestive suicidal overtones. These events were not

306


consciously thought of as suicide attempts by the adolescents; instead, they simply described their actions as a decision to continue shooting until no longer capable of doing so or until they were dead. Although guilt operated in these adolescents, it was often misattributed by professionals to some overt misdeed. Rarely did the clinicians connect the guilt as leading to provocative actions that, in turn, led to a cycle of attack and retaliation. In some cases, the cycle got out of hand and led to violence. One interpretation was that the adolescents’ capitulation to violence, such as in a group assaulting a victim, was a final act of despair to relieve a depressed state. These adolescents were frequently described by court personnel as being “callous” because they did not seem upset after committing a homicide. This callousness was especially noticeable when they declared that they felt better than they had before committing the homicide. The dynamic of inner relief related to the violent act, which temporarily displaced the depression, was often missed in assessments. Another model for interpreting this callousness was to interpret their calmness as a sense of relief that they seemed to experience after abandoning efforts to change their lives. To give up by participating in an act of violence meant that they could bypass demands to measure up to exaggerated expectations through an inflated ego ideal. Such giving up often seemed incongruous to those later involved in evaluating these cases, let alone to the public. That an adolescent would feel a homicide had been committed, but that everyone should now simply put it aside so they could get on with their lives, often evoked strong countertransference feelings in those who had to deal with them. These adolescents had minimal insight into the unreality of their desire to have the event simply put aside or forgotten. Because their depression was (even temporarily) alleviated, it was their belief that there was no need to dwell on the past. As one of them stated, “What’s done is done. Let’s get on with life.” Such verbalizations in a juvenile court system often resulted in their being labeled as psychopathic and promoted their certification to adult criminal court for trial. Although in some cases the adolescents had experienced excessive physical punishment to the point of physical abuse, in a number of cases an opposite phenomenon was observed. Distinguishing actual abuse from contrived abuse became problematic in many cases once a homicide had been committed. In these cases, the youths described being in a state of morbid physical fear of a parental figure. Yet by their own description, no beatings had occurred. One boy, later charged with murdering several members of his family, made recurrent statements involving his past fears that his father would beat him. Not until a de-


307

tailed psychiatric inquiry was made did it become clear that no such beatings had ever occurred. A similar distortion occurred involving a girl who alleged sexual abuse. Inquiry regarding episodes she described revealed that they had occurred 10 years earlier, when she was age 5 years, and had only involved her getting into bed to cuddle between her parents, something she remembered as being exciting. These descriptions were not conscious dissimulations by the juveniles, because they did not lie about the events. Rather, they were asserted as being factual events that had occurred, even though they had not. Clinicians evaluating adolescents must consider the phenomenon of false memory syndrome as a possible explanation for their assertions, especially if the adolescents had previous contact with therapists or evaluators who started the therapy or interview with assumptions about abuse. In some cases, the adolescents engaged in exaggerated hysterical behavior to portray themselves as being victimized, behavior that was related to repressed components in their personalities. In other adolescents, the distortion was related to their anger at realizing how dependent they were, with that realization giving rise to hatred. The anger was then projected onto parental figures, who were seen as persecutors. Often juveniles would use their familiarity (either personal or peer knowledge or knowledge gained through the media) with other abuse cases to provide legal justification for their actions in response to alleged abusive behavior. Another possibility encountered was continuation of a type of conflict that had commenced with the onset of puberty, a stage at which integrations and object choices were necessary but had not been accomplished. The naive quality in some adolescents’ exaggerations suggested conspicuous struggles with autonomy, often when flight occurred into involvement with a peer group to show their autonomy. At the same time, someone else would be blamed for their states of anxiety and depression. Retrospective evaluation of the 44 youths indicated activated sadomasochistic trends. The victims of the homicide were either loved ones or coincidental victims (i.e., those on the receiving end of displaced and projected aggression) who happened to be available on the right occasion at the right time. The perpetrator’s distorted process of thinking led to a failure of his or her defenses to contain the aggression. Intriguing questions arise about the dynamics of youths who, involved in even less favorable circumstances and possessing similar vulnerabilities, did not succumb to a homicide. Specificity in explaining why some do and others do not commit homicide is still lacking. In some situations involving depressed juveniles, a potential for homicidal behavior is reached inadvertently. Prime situations for homicides

308


are muggings, robberies, and encounters related to addictive drug use or distribution, among other situations. The following two case examples illustrate how easily these situations can eventuate in a homicide.

Case Examples A smaller boy with chronically low self-esteem and depression was baited by a physically stronger boy, which led to a confrontation in a parking lot. At one point, the stronger boy was pulling the smaller one by his hair to a place where the stronger boy wanted to continue beating him. The smaller boy eventually pulled out a knife and, with a single stab to the heart of the bigger youth, ended the fight. An adolescent boy would handle his depressed states by drinking. While drinking quite heavily, he would often become aware of a desire to fight someone. He would often frequent bars or areas near bars in his search for a fight. One night he selected a victim who was more intoxicated than he. They went into an alley, where only one blow was needed to end the actual fight. However, the youth then had an urge to keep beating the unconscious victim and proceeded to do so, including jumping on his head.

Neither of these cases started with a conscious wish to perpetrate a homicide, although presumably the adolescent in the second case eventually came to that point after the fight had begun. What about the depressed adolescent who becomes aware of a conscious desire to kill someone? One relevant question would be to inquire into the impairment of his or her ability to perceive reality and react correctly. Although this desire to kill may not reflect psychotic thinking to the degree of being prompted by a delusional system in which a supposed persecutor must be eliminated, it could signify a distortion in thinking to a degree that the potential perpetrator cannot assess his or her need to complete the act or appreciate the meaning or consequences of the act. The person usually does not entertain the possibility of his or her own destruction or death. A dominant adolescent may persuade a close friend or a romantic partner to assist in a parricidal act. A perplexing question is why such associates in these situations agree to participate. These participants seem to feel shame if they choose not to take part, like they are not measuring up to a distorted sense of duty or superego demands that they “come through.” In some of these cases, an adolescent originally may have agreed with a friend to help in killing the friend’s parents and then backed out a short time before the act. He or she then maintains this private knowledge, but his or her earlier role is revealed when the crime unravels. His or her sense of autonomy in making such a choice may


309

not correspond to a legal view in which he or she is charged as an accomplice or with conspiracy to murder. Depressed adolescents entertaining thoughts about homicide may have some factors in common with suicidal adolescents. One similarity is their belief that the state of despair they are in will continue without relief. A corollary belief is that the act of homicide will relieve their anxiety and tension. In retrospect, some type of event representing a personal failure with a resultant loss of self-esteem can often be identified as precipitating the homicidal act. Although such stressors seem to function to precipitate suicide in those with substance abuse problems, they also operate in depressed, homicidal adolescents. In the following case example, a girl’s failures to separate from a close attachment to her mother and the girl’s involvement in high-risk activities led to being an accessory to murder.

Case Example An adolescent girl had run away from home several times, during which she would work as a prostitute. She would then return home to her mother, who had strict religious beliefs and would emphasize the sinfulness and guilt associated with the girl’s actions. After one failed attempt at remaining separated from her mother, the girl once again planned to return home. Shortly before she intended to return home, she joined a pimp and another girl in “punishing” a third girl who was trying to escape. The “punishment” was for the girl to have her throat slit (this case was noted earlier). Following this event, there was no need for the girl to either return home or stay in the life she had been leading, for her participating in the homicide had put her fate into the hands of others. She had fought against dependence on her mother but could not deal with the hostility that accompanied such dependency. She recalled past fears that she would lose control and kill her mother.

Just as in a suicide, depressed adolescents feel trapped in a conflict from which they cannot extricate themselves by decisions of their own. In many cases, the transformation from a state of helpless passivity into one of taking charge of themselves becomes clear retrospectively. A variation on this theme is the Götterdämmerung finale. At this stage, the adolescent becomes convinced that nothing matters. In the suicidal state, this finale is reached when an individual’s ideation is turned into a belief that the world would be better off without him or her; in a homicidal state, this point is reached when the potential perpetrator comes to believe that nothing has gone right and the time has come to externalize the hopelessness within him or her. Homicidal individuals in this state are inhibited in their ability to think about the misery and suffering that will result for many other people from their

310


act or to feel guilt from their narcissistic absorption. At that point, even before the act has occurred, the question about who is to die is merely theoretical because the decision to kill has been made. The perpetrator experiences a sense of relief or calmness, given that the act is viewed as inevitable. Within this type of thinking, a homicide-suicide attempt may occur when the anxiety resulting from the realization that he or she is going to kill breaks through, which leads to a suicide attempt. In those adolescents whose homicidal efforts have somehow been aborted or called off, it is not infrequent for them later to minimize what happened. It may be laughed off as “just a joke.” These types of revelations require astute clinical judgment because at the time of revealing these thoughts, the patient’s homicidal impulse may be transiently gone. Yet these individuals still retain a sense of control from knowing that they can always put the plan into action if need be; this realization gives them a sense of power that allows them to view their future as being less bleak than it was before. Their sense of comfort is thus based on the power to effect a homicide when needed and not on any resolution to their conflicts. Unless the clinician can tap into this type of thinking, the difficulty is that these individuals may externally look quite intact and “talk a good game.” These types of maneuvers are encountered more with depressed adolescents who also have borderline personality features. The practical problems that emerge when the clinician becomes privy to such internal thought processes are enormous, especially the problems in obtaining the treatment such adolescents need. These adolescents’ fear of involvement in another relationship that may make them dependent, which is what they see themselves as having conquered, provides a tenuous basis for maintaining a treatment relationship. A clinician can easily join the adolescent in seeking to blame others as the source of difficulties, especially because an adverse environment often does exist around the adolescent. This mutuality between a clinician and adolescent is a treacherous alliance that carries a homicidal risk. One of the primary goals in treating these juveniles is to allow them to appreciate cognitive ways to gain control over their impulses other than simply blaming others or using the specious and grandiose ways in which they have been operating. The potential to regress and test out whether they are still in control always looms. The pressure from many sources will, unfortunately, be to bypass or ignore the more subtle aspects of their personalities that contribute to homicide, leading to the conviction that homicide or suicide is the answer. The adolescent, and many others involved, may push to forget the past or minimize it. Efforts are


311

often made to persuade the clinician that the homicide or homicidal thoughts were just a temporary aberration and should be forgotten. Such efforts at persuasion by the juvenile need close scrutiny.

SUBSTANCE ABUSE DISORDERS OR INTOXICATION Just as with adults, the use of psychoactive substances can be an important variable with homicides committed by juveniles. Whether the substance abuse is a primary diagnosis or secondary to some other primary diagnosis, psychiatric morbidity is the problem with these dual-diagnosis cases. When dealing with delinquent adolescents, some of whom are becoming more overtly violent as they age, one can see alcohol or drug abuse and intoxication patterns that began at quite an early age. Depending on the particular time and culture, different types of drugs begin to be used. Although many clinics report difficulty in controlling drug usage if another psychiatric diagnosis is seen as primary and gets treated, other clinics report the opposite—that is, that if the underlying condition is not treated, the drug and alcohol use continues. In reality, both conditions need treatment. Several studies have noted substance dependence problems in youths who have committed homicide. In a controlled study of 101 juvenile murderers with an equal number of nonviolent delinquents, 45% of the former abused alcohol compared with 28% of the latter.64 In another study of 18 juveniles who had committed murder, half were substance dependent.65 Similar findings are present in England, where 75% of 20 juvenile murderers abused alcohol, and in Finland, where 10 of 13 such juveniles were seen as dependent on alcohol.66,67 Because substance use may be heavily peer influenced or an attempt to deal with all manner of psychiatric problems, it is difficult to dissect conclusively the diverse interacting factors that lead to it. The progression from antisocial behavior to drug use and abuse and then to depressive disorders has been noted.68 The only remedy is to obtain a specifically detailed history and clinical examination of the individual in question. Even then, a separation of diagnoses may be difficult. Two complications may ensue in the treatment of dually diagnosed juveniles. One is the orientation and treatment philosophy of some drug treatment centers for juveniles, which view the use of any medications as unacceptable. This philosophy prevents the use of appropriate and controlled medication to treat psychiatric disorders other than those of substance use or dependence. The second is the viewpoint of drug abuse as the primary, if not sole, problem that has to be dealt with; this approach thereby may ignore other disorders within an individual personality or conflicts within the family.

312


A related question is whether there was intoxication at the time of the homicide. Some studies indicate a high percentage of juveniles on some substance at the time of the homicide. There appears to have been an increase in this group over time among the few studies. A 1977 California study found 25% of 31 juvenile homicide offenders were under the influence of drugs or alcohol at the time of the homicide.69 A 1987 study found that 38 (53%) of 72 juveniles who had killed were intoxicated.70 A New York study of 16 juvenile homicide offenders ages 16–17 noted 8 were “substance affected” at the time of the homicide, although they had been using lighter amounts than adults, which suggested the substances were having a greater effect on them.71 Unfortunately, none of these studies, in contrast to some with adults, noted whether the victims were also substance affected at the time.

NEUROPSYCHIATRIC CONTRIBUTIONS One of the more persistent, if not missed, areas in assessing violence in juveniles is the impact of neuropsychiatric deviations. This perspective originated some time ago in what were called the sequelae of prenatal and natal birth insults. Subsequent phenomena may be detected as well, such as various types of illnesses, injuries, exposures to toxins, or accidents that could have had an adverse impact on the developing central nervous system of a child. Frequently, adolescents with a tendency toward violent behavior may have a history of episodes of dizziness or blackouts, sometimes exacerbated by the use of alcohol or drugs. An earlier diagnosis of ADHD with concentration problems may complicate matters further. Neurological evaluations often come back with a lack of specific findings except for reports of various soft signs (e.g., poor coordination, spotty performances on memory tests, difficulty in fine motor coordination). The results from electroencephalographic tests are often of the diffuse generalized abnormality type, which leaves an unsatisfactory explanation. There may also be a correlation with underarousal in the central or autonomic nervous system. In one study, psychophysiological measurements at age 15 years appeared related to a status of criminality at age 24 years.72 This relationship could be genetically determined or caused by some developmental injuries. Related to this finding are questions about the possibility or significance of complex partial seizures (discussed in Chapter 2, “Biological Factors in Homicide”). Such considerations may infiltrate into the realm of possible neuropsychiatric disabilities. Two difficulties accompany these questionable neuropsychiatric findings: on one hand, the findings


313

are rejected because their lack of specificity does not allow one to draw any inferences from them; and on the other hand, combinations of suggested neurological findings are accepted as impacting the child by ultimately making him or her more unstable and by causing impulse control problems—a presumption that could accentuate the child’s difficulties. An ongoing debate continues about the significance of diverse neuropsychiatric findings and their contribution to homicides. Difficulties reside not only in assessments but also in interpretations of any positive findings. Related to the neuropsychiatric findings is the large number of juveniles displaying antisocial behaviors who have a low intelligence level. Although most are not severely retarded, many of them are in the borderline or mildly retarded group, just as their adult counterparts are. A large number in this group manifest early learning disabilities that perhaps are related to some of the suggestive neurological findings noted. Academic difficulties are not an infrequent finding when reviewing materials. The combination of all these factors—neuropsychiatric, intelligence level, and learning problems—in conjunction with the less privileged socioeconomic backgrounds of many of these youths places them in a high-vulnerability group for antisocial behavior. It is not an unwarranted inference to see that some of their impulsivity, difficulty in reasoning, and exercise of poor judgment over their behavior in many spheres are related to these antecedent disabilities. It is also well known that children with language and reading disorders have much more difficulty expressing their wants and needs as well as dealing with their feelings. From this perspective, they are much more likely to become impulsive and restless and to act out their feelings rather than develop ego controls and patience to deal with frustrating situations. Given these factors, gang participation, with its peer group and identity outside the conforming culture, would have great appeal. The drug culture, with its attendant risks of violence, further meets the needs of these juveniles, many of whom have preexisting psychosocial disabilities. As such, these factors, in a summative way, contribute to the possibility of a homicidal adolescent.

DISSOCIATIVE PHENOMENA The diagnostic category of dissociative disorders is often overlooked in adolescents, although theirs might be expected to be a prime age group for the manifestations of various types of dissociative phenomena. After

314


all, this diagnosis was originally labeled as hysteria and grouped with conversion disorders. Some adolescents who commit a homicide simply may have lapsed into a brief dissociative state, whereas a small number of these adolescents may actually fit the criteria for the controversial diagnosis of what was once referred to as a multiple personality disorder. In many respects, the diagnosis is overused and overdramatized. Whether dissociative identity disorder (DID) is a valid scientific entity, a figment of distorted imagination, or an iatrogenic state is not yet known despite ardent advocates for all positions. It is important to keep in mind that DSM-IV-TR has four other categories of dissociative disorders besides DID: dissociative amnesia, dissociative fugue, depersonalization disorder, and dissociative disorders not otherwise specified. Because the hallmark of these disorders is an alteration in consciousness that affects memory and identity, it would seem possible that one of these states could occur in the context of committing a homicide. Alteration in consciousness through either a constriction or expansion, which blurs boundaries between the self and others, is a phenomenon described in such situations. A complicating factor is that dissociative phenomena may occur with several other disorders, such as affective disorders, borderline personality disorders, and schizophrenia. Amnesia is another complicating factor with dissociative disorders. Some argue that the dissociative episodes themselves are related to some previous traumatic event that has been repressed.73 Others argue that they are not able to find evidence for amnesia for a traumatic event, apart from some organic factor or part of normal memory processes.74 These issues may get enacted in legal proceedings when there is an assertion that a defendant, juvenile or adult, is amnesic for events of a charged homicide. Legal questions regarding competency to proceed to trial may arise, as well as the relevancy of the condition to responsibility questions, such as the alter ego having committed the murder and therefore the act actually having been committed by “someone else.” When faced with these claims, courts, attorneys, and clinicians may start to grope to recover the memory or to rule out an event that occurred, and techniques are employed such as hypnosis or amobarbital interviews. The case of Andy Williams is instructive.75 He was a 15-year-old who walked into Santana High School in California in 2001 and shot 13 students and 2 staff members, which resulted in two deaths. He was described as quiet and nice, which left everyone mystified. He later described the feelings he had during the shootings to Diane Sawyer during a television interview: “I don’t think crazy is the right word. It’s, like, an out-of-body experience—when I was in my body, I was out of


315

my body at the same time.... I didn’t feel like it was actually me doing it.” The clinical mechanism hypothesized is that the dissociation is linked to violence through an underlying cycle of violence in which a murder could occur. In most of these cases, the individual is not denying that the act of murder occurred but rather is confused or perplexed. However, others think this is a matter of malingering. Because the individual may be claiming amnesia for the event but is not denying that it happened or that he or she did it, the question is whether such altered mental states should simply be ignored or be given some type of consideration.

ROLE OF CHILDHOOD ABUSE To those with a developmental or psychodynamic approach, it is not surprising that childhood experiences are viewed as significant in predisposing for later psychopathological effects that continue into adult life. This viewpoint has been reflected in some epidemiological studies.76 Because there are about 3 million reports of child abuse or neglect per year in the United States and about 1 million of them are substantiated, when a homicide occurs by a juvenile, the hypothesis of the significance of any abuse related to the act arises.77 This hypothesis is not a new one. In the early 1960s, Curtis78 wrote that abused and neglected children are tomorrow’s murderers and perpetrators of violence. From that point on, different clinical reports and research efforts have produced a mixture of impressions and conflicting findings. The specific question is whether such maltreatment can lead to serious violence, such as homicidal behavior, in adolescence and beyond. However, because the issue often arises with retrospective assertions after an act of violence, specifics about the impact of abuse are needed. Widom,79 in a large sample involving physical and sexual abuse cases, found that overall, children who were abused and neglected had higher rates of adult criminality than did a control group as measured by arrests for violent offenses. However, in another study, that same author did not find that abused or neglected children had higher rates of arrest for violent crimes as juveniles compared with control subjects, although females did appear to be at increased risk of arrest for violent crimes as adolescents.80 The conclusion was that empirical evidence for a “cycle of violence” could not be confirmed in terms of those who were abused recycling the abuse themselves. However, some research supports a link between the long-term consequences of physical abuse and adolescent aggression, if not homicidal

316


violence. Results from some studies have indicated that adolescents with violent aggressive behaviors that are displayed outside their families have a higher incidence of having experienced physical abuse than the general population.81 Other studies do not confirm this finding.82 A second research methodology has focused on adolescent boys in residential treatment facilities; these youths have shown a higher rate of past physical abuse compared with less violent or nonviolent male control groups. A third approach has focused on children receiving treatment for emotional problems. It was found that those who had been physically abused showed more aggressive behavior than a nonabused group.83 It must be remembered that the majority of abused and neglected children do not become delinquent or criminal, let alone engage in violent behavior. Mones,84 a defense lawyer who has defended 100 cases of parricide in court, argued that although even the most severely abused children do not necessarily kill their parents, in the cases where such homicides did occur, the children had all been abused. Yet of the estimated 1 million cases of substantiated child abuse per year, and approximately 16,000–18,000 homicides in the United States, only 300 involve murders of parents by children. To a psychiatrist, the most intriguing questions remain unresolved: What allows the homicides to take place when they do? Why do they not occur in at least a majority of these abused adolescents? The question remains, given conflicting research data, whether a significant subgroup who is abused then end up more prone than others to violence or to repeating abusive behavior. Some recent work suggests the possibility of genetic vulnerability for those later prone to the possibility of violence. Twins at a low risk for conduct disorder and who had experienced abuse showed a 2% increase in the probability of developing conduct disorders; among twins who were at high genetic risk, there was an increase of 24% for conduct disorders.85 Another study found that the monoamine oxidase A (MAO-A) gene can moderate the impact of early childhood maltreatment in the development of antisocial behavior in males.86 Genetic deficiencies in MAO-A activity have been linked with increased aggression and likelihood to develop antisocial behaviors. The question remains whether abuse is actually the most significant variable that contributes to the increased incidence of severe violence in adolescence. Another possibility is that abuse is one significant variable among many for those who later kill. Again, more specificity is needed as to why a subgroup behaves in a homicidal manner. Inconsistencies in the results of studies can be attributed to diverse study designs, reliance on reports of abuse, and use of arrests as a measure of violence.87 The debate continues.


317

There have been some control studies, which are difficult to conduct when investigating a rare event such as adolescent homicide. Busch et al.30 compared 71 homicide perpetrators in a group of 1,956 adolescent delinquents with 71 nonviolent delinquents. The homicide group was more likely to have a criminally violent family member, gang membership, an IQ less than 70, and alcohol abuse. Another study compared 21 males in prison who had been convicted of a murder before age 18 with a group of 21 convicted nonviolent burglars.88 The results showed “harsh parenting” from both parents in the murderer group, interpreted as an absence of protective parenting. Harsh parenting from the mother was seen as making a larger contribution than from the father. The socioeconomic status of the murderer group was lower than that of the burglary group. The pattern described for the group who had murdered was one of poverty; neglectful and violent parenting predisposing to early conduct disorder; and poor school performance leading to a downward spiral of failure, disruptive behavior, and exclusion from school and any experience of success. Lewis89 proposed a hypothesis to explain why maltreatment in the form of abuse or neglect may exacerbate preexisting psychobiological vulnerabilities. She viewed abusive behavior as leading to impulsivity and irritability, although whether as states or traits is not clear. The impulsivity and irritability engender hypervigilance, paranoia, diminished judgment capacity, and verbal incompetence. All of these consequences serve to curtail the recognition of pain in oneself and others. There is also the related question of innate temperamental differences. Although there is no specific finding that any genetic abnormality by itself predisposes one toward violent behavior, there is evidence that certain types of genetic characteristics affect the way an individual adapts. Genes are seen as affecting normal variations in attitudes and behavior in addition to disorders, and these dimensional effects may play a major role in psychopathological development.90 Differences in risk exposure or susceptibility to risk environments operate. Adaptations then interplay with the type of stressors in one’s environment. Biological components are often hypothesized, such as testosterone contributing toward vulnerabilities toward violence, which would fit the observation of more males being predisposed to homicidal behavior. Developmental and psychiatric evidence for the diverse sequelae from maltreatment is related to psychiatric problems and antisocial behavior. Many of these liabilities stem from difficulties in regulating behavior and relating to others. The adverse developmental sequelae related to abuse that pose increased risks for later violence may be seen in several areas:91 affect regulation; attachment relationships; develop-

318


ment of a sense of an autonomous self; difficulties in peer relationships in the form of aggressiveness, meanness, disruptiveness, and fewer prosocial behaviors; difficulties in adapting to school and risk of academic failure; and neurobiological sequelae in which a negative impact on the developing brain structure, function, and organization can give a distorted experience of the world. Perhaps as interesting a question as these negative factors is why some juveniles exposed to such adverse developmental experiences are still able to have a successful outcome. Those who survive such adverse experiences and have a successful outcome are often said to possess the quality of resilience. However, this seems to be a circular position because it refers to positive attributes, such as possessing good self-esteem and inner controls, as being the hallmarks of resilience. Yet the question is what distinguishes those who have these attributes, or can retain them despite abuse, from those who do not have these qualities. Where resilience occurs must be noted (i.e., assessing the types of risk situations where resilience is called upon), as well as with whom the juvenile is resilient—because resilience does not occur in a vacuum.92 Neurochemical response patterns that occur in the regulation of extreme psychological challenges also have to be considered.93 These patterns may provide more insights into the question of who has resilience and why they do.

SCHOOL SHOOTINGS The phenomenon of adolescents, or even preadolescents, carrying out multiple shootings in school settings has been dramatically portrayed in American society. It is often forgotten that violence in schools did not just develop as a source of concern in the late 1990s. In 1977, the federal government produced a report that found that teenagers only spent 25% of their time in school, but that 40% of robberies and 36% of physical attacks on teenagers occurred there.94 A 1999 survey of students ages 12–18 years noted that 2.5 million students were victims of crimes at school. Serious crimes accounted for 186,000 victims in schools (through rape, aggravated assault, sexual assault, and robbery), but 47 of these were school-associated deaths, which included 38 homicides.95 Even though fatal school shootings occupy a small number of cases of school violence, when they occur they are startling, especially because they often involve multiple victims. Firearms obviously increase the possibility of violence, and those who have been victims of crime, or who are in a peer group where weapons are carried, are most likely to


319

bring weapons into schools.96 Shootings accounted for 75% of schoolrelated deaths; stabbing/slashing 14%; beating/kicking 5%; and hanging, strangling, and unknown methods for the remainder, with 77% of the victims being males.97 The histories of diverse school shooting incidents could justify books for each event. The cases vary and are not simply duplicative acts, apart from the element of shootings. A few examples will have to suffice. The year before the famous Columbine High School shootings, a 13-year-old and his 11-year-old cousin in Jonesboro, AK, dressed up in camouflage clothing and set off a school fire alarm. The idea was to force students and teachers outside, where the boys could fire at them from a secluded wooded area. The boys used three rifles and seven handguns stolen from the homes of parents and grandparents. The result was 10 wounded and 4 girls and 1 teacher killed. The year before, a 16-year-old in Pearl, MS, killed his mother, then went to school and shot 2 students and wounded 7 others. That same year in West Paducah, KY, 3 students were killed and 5 wounded in a high school shooting.98 In 2005, the second-deadliest school shooting in U.S. history occurred when a 16-year-old boy on the Red Lake Indian reservation in Minnesota shot and killed his grandfather, who was a tribal police officer, and the grandfather’s companion, then put on a bulletproof vest and drove a police car to his high school. He then shot and killed 1 security guard, 1 teacher, and 5 students, for a total of 9 dead and 7 wounded. He then suicided. The Columbine High School shootings in Littleton, CO, in April 1999, have garnered a great deal of publicity. Eric Harris, age 18, and Dylan Klebold, age 17, killed 12 students and 1 teacher and wounded 20 students.99 The acts were methodically planned beforehand. The boys arrived at school that day with two 20-pound propane bombs and planted them in the cafeteria. Only one of them went off. Bombs were also placed in their cars to discharge when the boys were in the school, but they failed to go off. A bomb was planted a few miles away that exploded as a planned diversionary tactic. The first shots were fired in the school parking lot, where others reported them yelling, “Go! Go!” In these first shootings, 2 students were killed and 8 wounded. Harris and Klebold then entered the school and engaged in shooting exchanges with the police, who had arrived. A teacher who had spotted them was shot in the back and died. The two then ascended to the second floor, shooting on the way, and entered the library, where they killed 1 student. Survivors reported that the pair first taunted intended victims. Back in the cafeteria, Harris and Klebold tried to detonate the one failed bomb but it only caught on fire. They

320


ended up back in the library, where they exchanged fire with law enforcement before committing suicide by firing gunshots to the head. Investigation revealed that the pair intended to kill hundreds in order to be remembered as the greatest mass murderers of all time. One of the striking things about these types of juvenile shootings is how little we know about the perpetrators in terms of any psychopathology and adequate psychiatric knowledge. For cases like Harris and Klebold, information is only possible in an ex post facto manner. In other cases, there have been some preceding psychiatric and psychological evaluations, but the full reports are usually not released or subject to scrutiny or commentary. Much of the data gathered are often of the type law enforcement agencies use. Such data are valuable for that purpose but do not provide a psychiatric understanding. Information in many cases is limited to that provided by newspapers and the mass media, whose validity can be questioned. The wide coverage given to some of these crimes has also raised questions about a “copycat” effect. One approach used actions system theory in an attempt to understand school-associated homicides in terms of the interactions between offenders and their environments.100 The framework classifies the offenders within one of four operant styles: adaptive, conservative, integrative, and expressive. The adaptive offender targets specific individuals; the conservative is affected by an external trigger that threatens self-esteem; the integrative’s source and target for the killing is internal and the killing is often followed by a suicide; and the expressive is violent toward a random victim. A U.S. Secret Service study attempted to develop a profile of 41 school shootings.101 It is interesting that although the report concluded there was no evidence that the shootings were the result of a mental disorder, the perpetrators were described as feeling extremely depressed or desperate and that they had been picked on or bullied. The report described the most frequent motivation as revenge, and about threefourths of the perpetrators had threatened to kill themselves before the attack. Two-thirds of them described feeling persecuted, and threefourths were having difficulty dealing with a major change in a relationship or loss of status. The report noted that in 75% of the cases the shooters had communicated threats beforehand, sometimes in detail, whereas in other cases it was sensed that something was wrong by those in contact with them. The emphasis in the report was that the shooters had been bullied. The pervasiveness of bullying in schools leaves many unresolved questions about prevention and consequences for the bully and the bullied.102 Explanations for school shootings remain unsatisfying, and we are left with varying hypotheses. A Federal Bureau of Investigation (FBI)


321

report noted several “myths” that have emerged to explain such shootings, including revenge, anger about being bullied, unresolved anger about other matters, availability of guns in the homes of the perpetrators, the impact of violent video games, the shooter being a loner, and the shooters being similar.103 Perhaps seeking an explanatory model that would encompass all cases is a wrong approach. The similarities between a pair of 11- and 13-year-old shooters and a pair of 17- and 18year-old shooters might be minimal. Some investigators would bypass attempts at psychiatric explanations because they believe the shooters did not show signs of a mental disorder. Attempts may be made to create profiles or to carry out threat assessments of juveniles who pose a threat of violence to a target such as schools. School authorities or those in the community would be then put on notice about particular juveniles, but these approaches would seem to have minimal validity. The notices become so general and lacking in specificity that they are almost worthless. It has also been argued that a key variable when school shootings occur is the preexisting information about some of the juveniles and their dangerous potential that is not shared with others before the shooting.104 In one study of 253 school-associated violent deaths in the United States, more than half of the incidents had had some potential signal before the event, such as a note, threat, or journal entry.105 Missing some of the signals before shootings occur is due to the adolescent peer culture of not informing on others, but it is also due to the organizational structure of school systems, in which information about disturbed juveniles is lost or not communicated. A more recent FBI analysis of the Columbine shootings does away with several myths.106 These myths were that the shooters were targeting athletes and Christians, that the shooters were part of a “trench coat Mafia” group, and that they were outcast Goths. In fact, there were no specific targets, because the goal was rather to have the highest body count possible. The analysis concluded that the shootings were not impulsive, having been planned for a year. Although the slaughter was aimed at students and teachers, it was not motivated by resentment of them per se, except that they were the cornered quarry. Harris and Klebold had laughed about and resented the possibility of being seen as “petty school shooters.” The goal had actually been bombings and not merely shootings. If the bombs had been wired correctly, about 600 people would have become victims, thus achieving the greatest mass murder in United States history. The two were seen as different from one another; Klebold was hotheaded, depressive, and suicidal, whereas Harris on the surface was seen as “nice” but was actually cold, calculating, and homicidal—a psychopathic individual who took pleasure in

322


lying and was contemptuous of other people. The conclusion was that Klebold could never have pulled off Columbine if it were not for Harris.

TRANSFER TO THE ADULT JUDICIAL SYSTEM The disposition of juveniles charged with a homicide is a topic discussed in detail elsewhere.107 States vary in their procedures. Some states automatically certify juveniles to an adult criminal court if the juveniles are of a certain age, such as 16 or 17 years, and are charged with a homicide. Other states have a hearing to determine whether criteria are met to certify. Some jurisdictions have prosecutors file the case in adult court, with the burden then on the juvenile to get the case shifted into the juvenile court system. In the context of juvenile justice becoming increasingly punitive, several changes have occurred. One trend has been to lower the age in which certification to an adult court can occur; another is to lower the threshold for criminal culpability for adolescents. Although these trends have been criticized by some, others argue in favor of the deterrent effect of prosecuting juveniles in adult courts.108 The Byzantine character of diverse procedures in states has also been noted. When legal procedures of judicial waiver, direct filing in adult courts, and statutory exclusions are all considered as equivalent, it conveys an oversimplification of the judicial procedures that should be in place for dealing with juveniles. Similarly, to think that all juveniles charged with a homicide should be handled the same way because they have been charged with a similar legal act bypasses the need for any individual assessment.109 The significance of varying capacities of juveniles to participate in the juvenile and adult criminal justice systems raises serious policy questions.110 These questions are often ignored in practice. The question of developmental maturity was raised in terms of whether a person who committed a capital offense while he or she was a 17-year-old juvenile could be executed after being found guilty. Until 2005, those 15 years and younger could not be executed, whereas those 16 years and older were considered equivalent to adults in their thinking and could be executed. In the case of Roper v. Simmons,111 the U.S. Supreme Court decided in a 5– 4 decision to uphold a state supreme court ruling against executing juvenile offenders age 16 or 17 years at the time of committing a capital offense on the basis that it violated the eighth amendment’s prohibition against “cruel and unusual punishment.” This ruling means that the juvenile committing murder could still be certified and tried in adult court, but if he or she was found guilty of an offense punishable by execution, such as first-degree murder, an execution would be barred.


323

REFERENCES 1. Malmquist CP: Juveniles in adult courts: unresolved ambivalence. Adolesc Psychiatry 7:444–456, 1992 2. Grisso T, Zimring FE: Double Jeopardy: Adolescent Offenders With Mental Disorders (Adolescent Development and Legal Policy). Chicago, IL, University of Chicago Press, 2004 3. Arbuthnot J, Gordon DA, Jurkovic GJ: Personality, in Handbook of Juvenile Delinquency. Edited by Quay HC. New York, Wiley, 1987, pp 139–183 4. Flanagan TJ, Jamieson K: Sourcebook of Criminal Justice Statistics 1987. Washington, DC, U.S. Department of Justice, 1987 5. Snyder HN, Swahn MH: Juvenile Suicides, 1981–1998 (Youth Violence Research Bulletin). Washington, DC, U.S. Department of Justice, 2004 6. Centers for Disease Control and Prevention (CDC): Homicides among 15– 19-year-old males—United States, 1963–1991. MMWR Morb Mortal Wkly Rep 43:725–727, 1994 7. Bureau of Justice Statistics Clearinghouse: Ten Important Facts About Homicide: Homicide Statistics Information Package. Washington, DC, Bureau of Justice Statistics Clearinghouse, 1995, pp 148–462 8. Harms PD, Snyder HN: Trends in the Murder of Juveniles, 1980–2000 (Juvenile Justice Bulletin). Washington, DC, U.S. Department of Justice, 2004 9. McCurley C, Snyder HN: Victims of Violent Juvenile Crime (Juvenile Justice Bulletin). Washington, DC, U.S. Department of Justice, 2004 10. Cook PJ, Laub JH: The unprecedented epidemic in youth violence, in Youth Violence. Edited by Tonry M, Moore MH. Chicago, IL, University of Chicago Press, 1998, pp 27–64 11. Fagan J, Wilkinson DL: Guns, youth violence, and social identity in inner cities, in Youth Violence. Edited by Tonry M, Moore MH. Chicago, IL, University of Chicago Press, 1998, pp 105–188 12. Miller JG: Last One Over the Wall: The Massachusetts Experiment in Closing Reform Schools. Columbus, Ohio State University Press, 1991 13. Hamparian D: Youth in Adult Courts: An Introduction to Major Issues in Juvenile Justice Information Training (Readings in Public Policy). Columbus, OH, Academy for Contemporary Problems, 1981 14. Hamparian DM, Davis JM, Jacobson JM, et al: The Young Criminal Years of the Violent Few. Washington, DC, National Institute for Juvenile Justice and Delinquency, 1985 15. Farrington DP: Childhood aggression and adult violence, in The Development and Treatment of Childhood Aggression. Edited by Pepler DJ, Rubin KH. Hillsdale, NJ, Erlbaum, 1991, pp 33–61 16. United States Advisory Board on Child Abuse and Neglect: Neighbors Helping Neighbors: A New National Strategy for Protection of Children. Washington, DC, National Clearinghouse on Child Abuse and Neglect, 1993 17. Luntz BK, Widom CS: Antisocial personality disorders in abused and neglected children grown up. Am J Psychiatry 151:670–674, 1994

324


18. Smith C, Thornberry TP: The relationship between childhood maltreatment and adolescent involvement in delinquency. Criminology 33:451– 481, 1995 19. Wolfgang M, Figlio R, Sellin T: Delinquency in a Birth Cohort. Chicago, IL, University of Chicago Press, 1972 20. Strasburg P: Violent Delinquents. New York, Monarch, 1978 21. McDermott MJ, Hindelang MJ (eds): Juvenile Criminal Behavior in the United States: Its Trends and Patterns (Analysis of National Crime Victimization Survey Data to Study Serious Delinquent Behavior). Albany, NY, Criminal Justice Research Center, 1981 22. Rutherford A: Crime, law enforcement and penology, in Britannica Book of the Year. Chicago, IL, Britannica Book of the Year, 1994, pp 116–120 23. National Center for Health Statistics: Vital Statistics of the United States, 1989, Vol 2: Mortality, Part A. Hyattsville, MD, National Center for Health Statistics, 1992 24. Fingerhut MA, Kleinman JC, Godfrey E, et al: Firearm mortality among children, youth, and young adults 1–34 years of age, trends and current status: United States, 1979–88. Mon Vital Stat Rep 39(suppl):1–15, 1991 25. Centers for Disease Control: Weapon-carrying among high school students: United States, 1990. MMWR Morb Moral Wkly Rep 40:651–681, 1991 26. Webster DW, Gainer PS, Champion HR: Weapon carrying among innercity junior high school students: defensive behavior versus aggressive delinquency. Am J Public Health 83:1604–1608, 1993 27. Callahan CM, Rivara FP: Urban high school youth and handguns. JAMA 267:3038–3042, 1992 28. Cook PJ, Laub JH: After the epidemic: recent trends in youth violence in the United States, in Crime and Justice: A Review of Research. Edited by Tonry M. Chicago, IL, University of Chicago Press, 2002, pp 1–37 29. Loeber R, Farrington D: Young children who commit crime: epidemiology, developmental origins, risk factors, early interventions and policy implications. Dev Psychopathol 12:737–762, 2000 30. Busch KG, Zagur R, Hughes JR, et al: Adolescents who kill. J Clin Psychol 46:472–485, 1990 31. Shumaker DM, McGee GR: Characteristics of homicidal and violent juveniles. Violence Vict 16:401–409, 2001 32. Dolan M, Smith C: Juvenile homicide offenders: 10 years’ experience of an adolescent forensic psychiatry service. Journal of Forensic Psychiatry 12:313–329, 2001 33. Heide KM: Juvenile homicide encapsulated, in Juvenile Justice Sourcebook. Edited by Roberts AR. New York, Oxford University Press, 2004, pp 424–474 34. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000


325

35. Loeber R, Burke JD, Lahey BB, et al: Oppositional defiant and conduct disorder: a review of the past 10 years, part I. J Am Acad Child Adolesc Psychiatry 39:1468–1484, 2000 36. Burke JD, Loeber R, Birmaher B: Oppositional defiant and conduct disorder: a review of the past 10 years, part II. J Am Acad Child Adolesc Psychiatry 41:1275–1293, 2002 37. Kernberg PF: Personality disorders, in Textbook of Child and Adolescent Psychiatry. Edited by Weiner JM. Washington, DC, American Psychiatric Press, 1991, pp 515–533 38. Paris J: Personality Disorders Over Time: Precursors, Course, and Outcome. Washington, DC, American Psychiatric Publishing, 2003 39. Malmquist CP: Conduct disorder: conceptual and diagnostic issues, in Textbook of Child and Adolescent Psychiatry. Edited by Weiner JM. Washington, DC, American Psychiatric Press, 1991, pp 279–287 40. Cantwell CP, Baker L: Issues in the classification of child and adolescent psychopathology. J Am Acad Child Adolesc Psychiatry 27:521–533, 1988 41. Ben-Amos B: Depression and conduct disorders in children and adolescents: a review of the literature. Bull Menninger Clin 56:188–208, 1992 42. Kruesi MJP, Hibbs ED, Zahn TP, et al: A 2-year prospective follow-up study of children and adolescents with disruptive behavior disorders. Arch Gen Psychiatry 49:429–435, 1992 43. Halperin JM, Sharma V, Siever LJ, et al: Serotonergic function in aggressive and nonaggressive boys with attention deficit hyperactivity disorder. Am J Psychiatry 151:243–248, 1994 44. Hollander E, Stein DJ, DeCaria CM, et al: Serotonergic sensitivity in borderline personality disorder: preliminary findings. Am J Psychiatry 151: 277–280, 1994 45. Gurvits IG, Koenigsberg HW, Siever LJ: Neurotransmitter dysfunction in patients with borderline personality disorder. Psychiatr Clin North Am 23:27–40, 2000 46. Rapoport JL, Ismond DR: DSM-III Training Guide for Diagnosis of Childhood Disorders. New York, Brunner/Mazel, 1990 47. Hemphill JF, Hare RD, Wong S: Psychopathy and recidivism: a review. Legal and Criminological Psychology 3:139–170, 1998 48. Campbell MA, Porter S, Santor D: Psychopathic traits in adolescent offenders: an evaluation of criminal history, clinical and psychosocial correlates. Behav Sci Law 22:23–47, 2004 49. Mannuzza S, Gittelman-Klein R, Konig PH, et al: Hyperactive boys almost grown up, criminality and its relationship to psychiatric status. Arch Gen Psychiatry 46:1073–1079, 1989 50. Rutter M: Pathways from childhood to adult life. J Child Psychol Psychiatry 30:23–25, 1989 51. Loeber R, Stouthamer-Loeber M: Prediction, in Handbook of Juvenile Delinquency. Edited by Quay HC. New York, Wiley, 1987, pp 325–382

326


52. Ellickson PL, McGuidan KA: Early predictions of adolescent violence. Am J Public Health 90:566–572, 2000 53. Westen D, Shedler J, Durrette C, et al: Personality diagnoses in adolescence: DSM-IV Axis I diagnoses and an empirically derived alternative. Am J Psychiatry 160:952–966, 2004 54. Lewis DO, Pincus JH, Bard B, et al: Neuropsychiatric, psychoeducational, and family characteristics of 14 juveniles condemned to death in the United States. Am J Psychiatry 145:569–593, 1988 55. Benedek EP, Cornell DG: Juvenile Homicide. Washington, DC, American Psychiatric Press, 1989 56. Labelle A, Bradford JM, Bourget D, et al: Adolescent murders. Can J Psychiatry 36:583–587, 1991 57. Bender L: The concept of pseudopsychopathic schizophrenia in adolescents. Am J Orthopsychiatry 29:491–509, 1959 58. Clare P, Bailey S, Clark A: Relationship between psychotic disorders in adolescence and criminally violent behaviour. Br J Psychiatry 177:275–279, 2000 59. Lewis DO, Moy E, Jackson L, et al: Biosocial characteristics of children who later murder: a prospective study. Am J Psychiatry 142:1161–1167, 1985 60. Malmquist CP: Depression and extreme violence in adolescence, in Thinking Clearly About Psychology, Vol 2: Personality and Psychopathology. Edited by Grove WM, Cicchetti D. Minneapolis, University of Minnesota Press, 1991, pp 378–401 61. Hathaway SR, McKinley JC: Minnesota Multiphasic Personality Disorder, Revised. Minneapolis, University of Minnesota, 1970 62. Beck AT: Depression Inventory. Philadelphia, PA, Philadelphia Center for Cognitive Therapy, 1978 63. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC, American Psychiatric Association, 1980 64. Zagar R, Arbit J, Sylvies R, et al: Homicidal adolescents: a replication. Psychol Rep 67:1235–1242, 1990 65. Myers WC, Scott K: Psychotic and conduct disorder symptoms in juvenile murderers. Homicide Studies Journal 2:160–175, 1998 66. Bailey S: Adolescents who murder. J Adolesc 19:19–39, 1996 67. Santtila P, Haapasalo J: Neurological and psychological risk factors among young homicidal, violent and nonviolent offenders in Finland. Homicide Studies Journal 1:234–253, 1997 68. Rutter M: Substance use and abuse: causal pathways considerations, in Child and Adolescent Psychiatry, 4th Edition. Edited by Rutter M, Taylor E. Oxford, England, Blackwell Scientific, 2002, pp 455–462 69. Sorrells JM: Kids who kill. Crime Delinq 23:313–320, 1977 70. Cornell DG, Benedek EP, Benedek DM: Characteristics of adolescents charged with homicide. Behav Sci Law 5:11–23, 1987


327

71. Fendrich M, Mackesy-Amiti ME, Goldstein P, et al: Substance involvement among juvenile murderers: comparisons with older offenders based on interviews with prison inmates. Int J Addict 30:1363–1382, 1993 72. Raine A, Venables PH, Williams M: Relationships between central and autonomic measures of arousal at age 15 years and criminality at age 24 years. Arch Gen Psychiatry 47:1003–1007, 1990 73. Brown D, Scheflin AW, Hammond DC: Memory, Trauma Treatment and the Law. New York, WW Norton, 1998 74. Pope HG Jr, Hudson JI, Bodkin JA, et al: Questionable validity of ‘dissociative amnesia’ in trauma victims: evidence from prospective studies. Br J Psychiatry 172:210–215, 1998 75. Moskowitz A: Dissociation and violence. Trauma Violence Abuse 5:21–46, 2004 76. Kessler RC, Davis CG, Kendler KS: Childhood adversity and adult psychiatric disorder in the US National Comorbidity Survey. Psychol Med 27: 1101–1119, 1997 77. Spinazzola J, Ford JD, Zucker M, et al: Survey evaluates complex trauma exposure, outcome, and intervention among children and adolescents. Psychiatr Ann 35:433–439, 2005 78. Curtis CG: Violence breeds violence—perhaps? Am J Psychiatry 120:386– 387, 1963 79. Widom CS: The cycle of violence. Science 244:160–166, 1989 80. Widom CS: Childhood victimization and adolescent problem behaviors, in Adolescent Problem Behaviors. Edited by Ketterlinus R, Lamb ME. New York, Erlbaum, 1994, pp 127–164 81. Vissing YM, Straus MA, Gelles RJ, et al: Verbal aggression by parents and psychosocial problems of children. Child Abuse Negl 15:223–238, 1991 82. Fagan J, Hansen K, Jang M: Profiles of chronically violent delinquents: empirical test of an integrated theory, in Evaluating Juvenile Justice. Edited by Kleugel J. Beverly Hills, CA, Sage, 1993, pp 91–119 83. Malinosky-Rummell R, Hansen DJ: Long-term consequences of childhood physical abuse. Psychol Bull 114:68–79, 1993 84. Mones P: When a Child Kills: Abused Children Who Kill Their Parents. New York, Pocket Books, 1991 85. Jaffee SR, Caspi A, Moffitt TE, et al: Nature X nurture: genetic vulnerabilities interact with physical maltreatment to promote conduct problems. Dev Psychopathol 17:67–84, 2005 86. Caspi A, McClay J, Moffitt T, et al: Role of genotype in the cycle of violence in maltreated children. Science 297:852–854, 2002 87. National Academy of Sciences: Understanding Child Abuse and Neglect. Washington, DC, National Academy Press, 1993 88. Hill-Smith AJ, Hugo P, Hughes P, et al: Adolescent murderers: abuse and adversity in childhood. J Adolesc 25:221–230, 2002 89. Lewis DO: From abuse to violence: psychophysiological consequences of maltreatment. J Am Acad Child Adolesc Psychiatry 31:383–391, 1992

328


90. Rutter M: The interplay of nature, nurture, and developmental influences. Arch Gen Psychiatry 59:996–1000, 2002 91. Cicchetti D, Toth SL: Child maltreatment. Annual Review of Clinical Psychology 1:409–438, 2005 92. Bolger KE, Patterson CJ: Sequelae of child maltreatment/vulnerability and resilience, in Resilience and Vulnerability. Edited by Luthar SS. New York, Cambridge University Press, 2003, pp 156–181 93. Charney DS: Psychological mechanisms of resilience and vulnerability: implications for successful adaptation to extreme stress. Am J Psychiatry 161:195–216, 2004 94. U.S. Department of Health, Education and Welfare: Violent Schools—Safe Schools: The Safe Schools Study Report to the Congress, Vol 1. Washington, DC, U.S. Government Printing Office, 1997 95. Kaufman P, Chen X, Choy S, et al: Indicators of School Crime and Safety, 2001. Washington, DC, U.S. Department of Education and Bureau of Justice Statistics, 2001 96. Wilcox P, Clayton R: A multilevel analysis of school-based weapons possession. Justice Quarterly 18:509–542, 2001 97. National School Safety Center: School Associated Violent Deaths Report. Westlake Village, CA, National School Safety Center, 2005. Available at: http://www.nssc1.org/savd/savd.pdf. Accessed October 19, 2005. 98. Diane M: Too High a Price for Harmony: A Perspective on School Shootings. Bloomington, IN, First Books Library Publishers, 2000 99. Petee TA, Wittekind JEC: Columbine/Littleton school shootings, in Encyclopedia of Murder and Violent Crime. Edited by Hickey E. Thousand Oaks, CA, Sage, 2003, pp 86–87 100. Fritzon K, Brun A: Beyond Columbine: a faceted model of school-associated homicide. Psychology, Crime and Law 11:53–71, 2005 101. Vossekuil B, Reddy M, Fein R, et al: Safe School Initiative: An Interim Report on the Prevention of Targeted Violence in Schools. Washington, DC, United States Secret Service, 2000 102. Bond L, Carlin JB, Thomas L, et al: Does bullying cause emotional problems? a prospective study of young teenagers. BMJ 323:480–484, 2001 103. O’Toole ME: The School Shooters: A Threat Assessment Perspective. Quantico, VA, Critical Incident Response Group, National Center for Analysis of Violent Crime, 2000 104. Newman KS: Rampage: The Social Roots of School Shootings. New York, Basic Books, 2004 105. Anderson M, Kaufman J, Simon TR, et al: School-associated violent deaths in the United States, 1994–1999. JAMA 286:2695–2702, 2001 106. Cullen D: The depressive and the psychopath: at last we know why the Columbine killers did it. Slate, April 20, 2004. Available at: http:// slate.msn.com/id/2099203. Accessed April 20, 2004.


329

107. Malmquist CP: Juveniles and the adult criminal justice system: transfer/ waiver to adult court, in Textbook of Adolescent Psychiatry. Edited by Rosner R. London, England, Arnold, 2003, pp 489–494 108. Levitt SD: Juvenile crime and punishment. J Polit Econ 106:1156–1185, 1998 109. Fagan J, Zimring FE (eds): The Changing Burdens of Juvenile Justice: Transfer of Adolescents to the Criminal Court. Chicago, IL, University of Chicago Press, 2000 110. Schmidt MG, Reppucci ND, Woolard JL: Effectiveness of participation as a defendant: the attorney–juvenile client relationship. Behav Sci Law 21:175–198, 2003 111. Roper v Simmons, 543 US 551 (2005)


CHAPTER

10 SEXUAL HOMICIDE

THE TOPIC OF SEXUAL HOMICIDE is a general one that includes many types of homicide. Some of the victims of these killings are children, whereas others are adults. Some victims are chosen on a heterosexual basis, some on a homosexual basis; a third group of sexual homicides involves bisexual perpetrators who select different types of victims over time. In addition, some investigators believe that serial killers (discussed in Chapter 1, “Epidemiological Aspects of Homicide”) are all a variant of sexual killers; however, this theory is debatable and is a major source of confusion. It is more accurate to say that only some serial murderers are sexual murderers and only some sexual murderers are serial murderers. Complicating matters further is by what standards a killing is interpreted as being sexualized. Such an interpretation goes beyond seeing the acts in a purely descriptive sense because the distinction involves interpreting behaviors as sexualized, even when a killing has not involved overtly sexual acts. Perhaps the biggest source of confusion and error occurring in the literature is drawing inferences about sexual offenders in general and then applying the deductions to the subgroup of those who are sexual killers. Sexual offenders themselves compose a heterogeneous group, including rapists, pedophiles, exhibitionists, and voyeurs, to name a few. The great majority of these individuals do not kill and do not even physically assault others. It is unfortunate that most studies have not differentiated sexual from nonsexual homicides and, more specifically, have not differentiated various types of sexual killers. The complication remains in determining whether a particular homicide should be classified as a sexual homicide. Note that the Uniform Crime 331

332


Reports lists the number of murders per year in the United States but does not differentiate within the group which homicides were sexual homicides. Sometimes the Supplementary Homicide Reports help, but the complexity of motivations and diverse behaviors leaves the matter unsettled. For example, a homicide preceded by a sexual assault is simply classified as a homicide. Hence, accurate figures on the incidence of sexual homicide do not exist. Often a public focus on high-publicity serial sexual killers and their trials leaves a vacuum of attention on the other types of sexual homicides. In addition, a more subtle problem exists in determining whether a particular murder should be classified as a sexual homicide. Even a definition with face validity, such as stating a sexual homicide is based on sexual activity occurring during, before, or after, the course of committing a criminal offense, is so broad as to encompass many homicides not usually thought of as sexual homicides. In the discussion to follow, rape homicides and homicides to conceal evidence are homicides but not sexual homicides in the strict sense. Even in cases that appear to have the hallmarks of a sexual homicide, such as accompaniments of genital mutilation, bite marks, or “signature” signs, the act may have occurred in connection with some bizarre delusional thinking. Sometimes erotized fantasies are enacted but without overt sexual acts. For example, a woman is first kidnapped and tied up with rope as a hostage. She is not sexually molested and no ransom is sought. After several days she is killed. One viewpoint might view the killing as an effort to prevent the identification of the perpetrator. However, interview material from the perpetrator indicates that an urge to kill her became more powerful and in time overcame his resistance.

TAXONOMY OF SEXUAL HOMICIDE Sexual homicides may be classified operationally in different ways. A working breakdown could comprise the following: 1. Rape homicides 2. Sexual lust (or sadistic) homicides 3. Homicides after a sexual act to destroy evidence This taxonomic breakdown has the advantage of allowing diverse approaches to clinical or criminological theorizing and an integration in each of these categories, as well as applying individual psychiatric diagnoses in a descriptive or psychodynamic sense. This classification further allows developmental antecedents in the individual to be considered as well, something that is quite important in theorizing about the significant vari-

Sexual Homicide

333

ables that may be present and that contribute to such an outcome. The individual’s various significant or traumatic experiences, including his or her socioeconomic background, can be encompassed by this classification.

Rape Homicides Rape homicides are those in which a sexual act, and not the killing per se, is seen as the primary motive for the act; simply put, in the course of the rape, a homicide occurs. However, an important note is that in practice, this seemingly neat distinction often breaks down. The following are examples of this type of killing: • In the middle of a rape, the victim resists, and the perpetrator attacks the victim or strikes him or her. • In the middle of a sodomitic assault on a child, the child attempts to escape, the perpetrator strikes the child, and the child is killed. • In one case, a serial killer killed three women on three different occasions and a woman and her two children on another occasion; all these deaths occurred by strangulation. Over a period of years, the perpetrator had developed the need to strangle the victim during sexual intercourse if he was to ejaculate; in most cases, he would release the woman from the strangulating hold before she expired, with the release ideally occurring simultaneously with ejaculation. In the cases of the homicides, the ejaculation did not occur early enough and the women died of asphyxiation. Two of the women he killed were prostitutes. A third woman was one he had met in a bar and subsequently stayed with at her apartment for a few days while en route across the country to a different city. The fourth was a woman he had been living with for several months; he stated that he had killed her children after killing her to allow him extra time to leave the area without being discovered (the children’s deaths are an example of the third type of homicide—to destroy evidence). His later explanation of the acts was that the killings were simply accidental; he was unable to appreciate his compulsive need to have his sexual partners helpless and near death as a condition for his completing the sexual act. In fact, his masturbation fantasies involving necrophilic acts revealed this connection.

Lust Homicides Sexual lust or sadistic killers are sometimes seen as the true sexual killers. The phrase “lust killing” stems from the original work of Krafft-Ebing1 in his 1898 discussion of sadistic homicides. In contrast with the first group,

334


in which a fusion of sexuality and aggression sometimes directly, but often inadvertently, leads to a homicide, in lust killings the primary goal is to kill the victim as part of a ritualized attack. Although the killing sequence may include an act of sexual intercourse, sexual intercourse does not always occur, and other types of sexual acts may be part of the homicide. The primary motivation for the perpetrator is the enactment of some type of fantasy that has preoccupied him or her for some time. The nature of the fantasy may have changed over time. It is unknown how many individuals in society have such sadistic sexual fantasies but never enact the fantasies in actuality. The clinical speculation is that the number might be very high compared with those who actually do carry out some type of act. The primary question to be asked is, what factors must be operative to allow a subgroup of individuals to break through their defenses and act out this fantasy? Presumably, some alteration in ego controls must occur, but why it occurs at a given time is a persistent question. From a clinical standpoint, most likely a series of factors contribute to such a weakening of control; for example, often months or years of inner conflict, turmoil, and debate are present before the person commences such behavior, although there are exceptions. In this context, themes of power, dominance, and subjugation of a victim are key; when acts of mutilation or mayhem occur, they have been part of the prevailing fantasy. Rarely has such an act occurred in isolation without such a fantasized enactment.

Homicides to Destroy Evidence The third category of sexual homicides is not strictly defined as sexual killings. In these cases, some kind of sexual act occurs, and the victim (or observers, as illustrated in the case described earlier in the section “Rape Homicides,” in which the women’s children were killed) is murdered to prevent discovery. Except for the sexual act itself, these would not be classified as sexual homicides in a taxonomic schema.

DESCRIPTIVE MODELS OF SEXUAL HOMICIDE General Considerations Based on the State of the Victim From these introductory points, it becomes clearer that it is often difficult to determine what should be classified as a sexual homicide. As noted, some investigators believe that most serial killers are de facto sexual murderers. However, a killing in the context of a quarrel between lovers, or a spousal killing, would not be classified as a sexual homicide. Similarly, re-

Sexual Homicide

335

taliatory types of homicide in which there is a sense of betrayal by a loved one would not be viewed as sexual homicides but rather as crimes of passion. Therefore, the key question is, what are the essential hallmarks, even in a descriptive sense, for an act to be labeled a sexual homicide? Simple evidence of an act of sexual intercourse or anal or vaginal penetration having occurred in the context of a homicide does not in itself mean a sexual homicide has occurred, except perhaps in the very broadest sense. Rather, what seems to be the defining descriptive characteristic is evidence of a more bizarre or brutal aspect of a homicide that accompanied a sexual attack. The boundaries remain vague, but some more obvious cases would be the following: • • • •

A dead body being left in a sexually suggestive pose Objects having been inserted in body cavities (e.g., the anus, vagina) Mutilation of parts of the body, if not outright dismemberment The manner in which the body was clothed, covered, or left uncovered • Suggestive evidence of sadistic or masochistic practices that were carried out (e.g., use of restraints, attempts at asphyxiation, evidence of dehumanizing treatment of the victim) • Repetitive ritualistic details on successive victims, such as body positioning, mutilations, and so on, that are symbolic for a particular perpetrator (“signature” or “calling cards”) These are suggestive descriptive aspects from the external examination of the bodies of the victims and examination at the scene of the crime, independent of any knowledge about the perpetrator.

Federal Bureau of Investigation Study Investigative leads on the perpetrator compose another line of investigation into sexual homicides. In a research volume on sexual homicides written by a nursing professor and two supervisory agents of the Federal Bureau of Investigation (FBI), Ressler and colleagues2 studied 36 convicted and incarcerated sexual murderers. It should be noted that the data on these perpetrators were not all complete but that the majority of these men did at least agree to participate in the study. To date, this has probably been the largest group of sexual killers to be interviewed for research purposes; the authors’ research subsequently has been extensively summarized by others. All of the perpetrators were men and came from different geographical areas. Twenty-nine of the group had attacked multiple victims and 7 had had only 1 victim, for a

336


total of 118 victims. Nine victims survived; these cases were handled as attempted murder cases. The perpetrators were suspects in more cases than those recorded but were never brought to trial for these other cases. Official records were used as well as interviews with the offenders; thus the data primarily reflect the recall of the perpetrators. The offenders gave multiple reasons for participation. Some who admitted the offense stated it gave them an opportunity to contribute to an increased understanding about them. Those who denied guilt saw it as an opportunity to illustrate why they could not have committed the crime. Others saw it as an opportunity to “teach” the police how the crimes were committed and about their motivations. These types of conscious reasons undoubtedly concealed a diverse set of motives, but they are interesting in their own right. A law enforcement orientation to sexual homicides, as in the FBI study, has its strengths and weaknesses. The emphasis is on a classification system based on an analysis of the crime scene. As such, this orientation reflects what a good detective does when confronted with a homicide, such as focusing on details at the scene of the crime for purposes of apprehending the guilty party. Details about the position of the victim, the condition of the body, use of a weapon, and so on are basic to such an approach. This type of primary operational mission has the worthwhile goal of promoting the capture of sexual murderers at large and secondarily preventing other attacks from recurring. What it lacks is a type of in-depth psychiatric explanation and understanding of the murderers themselves, although it may provide suggestions about their developmental history. At best, such an operational approach can only suggest certain psychological and social variables that were also present in the sexual murderer’s profile.

Motivational Model Ressler and colleagues2 did propose a motivational model involving certain critical personal traits as well as cognitive mapping and processing that they believed interacted together to produce patterned responses. Their extensive data on this overall group of murderers offer the best set of descriptive items about a group of sexual murderers available. The FBI study began with a focus on the perpetrators’ early life attachments or their early bonding patterns, which offered a blueprint of how perpetrators began to perceive life outside their family settings. The social bonding process that occurs in individuals who commit sexual homicides is seen as either failing to occur adequately or being quite narrow and selective in its outcome. The adult caretakers who reared the future sexual murderers either ignored, rationalized, or tried to normalize the

Sexual Homicide

337

behaviors that emerged in the boys. From the adults’ own personal problems, the distortions and projections occurring in their children were given support. In effect, the study approach was a developmental theory dealing with the parental impact on a child in the presence of parental conflict or preoccupation and the resultant distorted impact these factors had on the development of the child. The study approach amounted to a theory with a developmental psychopathological perspective for the emergence of delinquent behavior in general. However, as a theory, it lacked specificity to predict an ultimate outcome of murderous behavior, let alone a sexual homicide. According to this theory, these boys were not able to register the fact that their antisocial behavior should be punished because the behavior had been “normalized” in these families by generalizations such as “all boys get into trouble.” This variable in family background is frequently encountered in many delinquent youths. To achieve more specificity, the authors proposed three factors that are formative events in the life of a child who has the potential for later becoming a sexual murderer: 1. The first factor was having experienced either direct physical or sexual trauma or some indirect trauma (e.g., witnessing family violence). In such a situation, the child is viewed as not receiving the type of protection needed to deal with the impact such events have on him or her. From these frightening childhood experiences, memories emerge in the context of daydreams, fantasies, nightmares, and disturbing memories of the events. The unsuccessful resolution of these traumas leaves the child feeling victimized and helpless. As a response, he or she begins to use aggressive fantasies to achieve a dominance and control that has been absent in the real world. The abusive event may also induce a sustained emotional-physiological arousal that interacts with the child’s thoughts about the trauma and alters his or her perceptions and patterns of interpersonal life. Adult perpetrators of sexual abuse thus have a higher incidence of childhood sexual victimization experiences in their backgrounds than nonperpetrators.3 However, this factor is one step along the way—but insufficient in itself—to explain a subsequent sexual homicide. The perception and meaning of the trauma and how it unfolds for the individual are needed for specificity. 2. The second formative event proposed was a developmental failure: the blocking of attachments that should have taken place between the child and his or her primary caretaker resulted in the adult lacking influence with the child. The authors did not offer any explanation of why this blocking occurred, although they acknowledged that they were not addressing any neurological or genetic factors.

338


3. The third hypothetical factor was the failure of an adult to serve as a role model for the child, such as a parent being absent or inadequately filling the parental role. It is unfortunate that the FBI study did not add a clinical dimension, for this could have provided other sources of knowledge about the individuals who committed sexual homicides. Thus the clinical questions as to how or why such developmental failures occurred, and what diagnostic possibilities were present that specifically related to later sexual murdering, remain unanswered. The motivational model proposed by these authors has several components. The first level is an ineffective social environment, and the second level is the formative events from childhood and adolescence. The third level, patterned responses, includes two subcategories: critical personality traits and cognitive mapping and processing. These subcategories are discussed below. The traits found in these men were related to their sense of social isolation. They would use fantasies related to autoerotic activities and fetishes as a substitute for human encounters. It is possible to argue that these men’s sexual development was restricted from the lack of mutual caring, pleasure, and companionship they had experienced. The isolation experienced by these men led to their having a sense of being different from others; however, they also expressed anger at a society that they viewed as rejecting them. Other traits seen in the men were rebelliousness, aggression, chronic lying, and a sense of privilege or entitlement that they should get what they wanted. Cognitive mapping and processing compose the other subcategory of patterned responses. These thinking patterns give control and development to the person’s internal life and link it to the social environment. Such mapping and processing give meaning to events and control the states of helplessness, terror, and pervasive anxiety encountered in the course of personal development. In sexual murderers, the mapping and processing are seen as fixed, negative, and repetitious patterns that give an antisocial view of the world. The fantasies then stimulate these men and decrease their tension. Their self-image is frightening, if not terrifying, to themselves and others because of the possibility that others could discover their fantasies. Themes of control and dominance over others become a substitute for mastery of their own internal and external experiences. The men in this study became aroused by high levels of aggressive experience that required stimulation. Themes of dominance, rape, molestation, power, torture, humiliation, inflicting pain on oneself or others, or death itself became the main focus of their fantasies.

Sexual Homicide

339

Another level in the motivational model involved taking actions toward others that were seen as an effort to achieve a joyless dominance. Childhood behaviors to achieve such dominance are manifested in being cruel to animals, abusing other children, engaging in destructive play patterns, setting fires, and committing various property offenses. If seen in a clinical setting, these juveniles would be given some type of conduct disorder diagnosis. By adolescence, more violent behavior emerges, such as assaults, burglary, rape, and the commencement of sexual assaultive behaviors. However, once again, this background description can fit many children who emerge into patterns of juvenile delinquency and perhaps later serious adult crime as well. This description lacks sufficient specificity to make predictions about who in such a group might actually emerge as a sexual murderer. The last level in the motivational model was a “feedback filter,” in which these men’s feelings of dominance, power, and control were increased. These feelings justified their earlier actions and helped them to sort out events to preserve the fantasies and avoid the external environment impinging on them, as well as to avoid detection.

Organized/Disorganized Dichotomy Another facet of the FBI study in developing a psychological profile was to separate serial killers into an organized/disorganized dichotomy that was to differentiate sexual homicides.4 Analogous to fingerprints used in ordinary crime scene investigations, the profile of an organized or disorganized killer attempted to help identify the perpetrator. The organized offender was described as an orderly person, which was also assumed to be the way he would then commit crimes. Thus, he would be seen as intelligent, have a skilled job, be verbally and socially skilled, and be reacting to some precipitating event. The offenses would be well planned, restraints would be used, and a weapon would be used for the murder (which would not be left at the scene). In contrast, the disorganized offender suggested minimal planning and someone of below-average intelligence who was socially incompetent. Varieties of evidence would be left at a chaotic crime scene, such as fingerprints, blood, semen, and perhaps the murder weapon, with only occasional use of restraints. The body would be poorly concealed if at all. The homicide would be viewed as opportunistic and be near the residence of the perpetrator. The study did provide an “out” by offering a third, “mixed” category, based on the possibility of some unforeseen factor that then did not permit a sexual homicide to be identified as either organized or disorganized.

340


An investigative group, Canter et al.5, pointed out how both crime investigations and academic writings have been influenced by the dichotomy approach. Using problem scaling, the group critically assessed the impact of this typological approach. The researchers found the assumption that the characteristics that define one type of killer always co-occur and do not frequently occur in crimes assigned to any other type. Based on analysis of cases tested by multidimensional scaling, the study found a co-occurrence of 39 aspects of serial killings. The results did not reveal any distinct subsets of offense characteristics reflecting a dichotomy. Rather, the results revealed a subset of organized features typical of most serial killers rather than one typology.

Other Descriptive Studies of Sexual Killers It is easy to assume that patterns of sexual aggression for more minor sexual offenses are the same as those present during sexual homicides. Such patterns would seem more likely for homicides that emerge during the course of sexual assaults. Whether the pattern also fits lust murderers is more problematic. To assume the same set of factors and fantasies for all sexual offenses leaves unresolved questions about why sexual murderers do not become even more prevalent than they seemingly are. Perhaps increased publicity of cases seen recently will allow researchers to accumulate more knowledge about these killers. Unfortunately, because of the limited number of cases, researchers’ knowledge base is still often focused on the few high-publicity cases, such as Jack the Ripper, Ted Bundy, David Berkowitz, or Dennis Nilsen. More recent notable cases include Gary Ridgway, the Green River Killer who confessed to killing 48 women in the Seattle, WA, and Portland, OR, areas over 16 years and who stated that strangling young women was his “career.” 6 It is believed that he murdered many more. Other notable cases have been Arthur Shawcross, the strangler of 11 women around Rochester, NY; Kenneth Bianchi in Los Angeles, CA; Jeffrey Dahmer in Milwaukee, WI (murdered in prison after conviction); and Joel Rifkin around New York City. Here is the clinical question, then, for which there is no adequate answer: when do the fantasies and enactments related to more minor sexual deviancies extend or shift to those involving more ominous sadistic, if not murderous, components? In one study of murders in England, 14% of the 306 murderers whose victims were females older than age 16 years were classified as being “sexually motivated.”7 However, the authors of that study had previously screened out those legally classified as insane or having a diminished-responsibility legal status. It could be presumed that the fig-

Sexual Homicide

341

ure would be higher in the United States, given that the overall rates of violence and homicide are higher. Many of the case reports appear to indicate that these individuals have severe disturbances in their personality functioning in a clinical sense. Many of these individuals seem to be in the aggressive sociopathic type of group. An early attempt at developing a profile of sadistic sexual murderers was done by Robert P. Brittain,8 who was both a forensic pathologist and a psychiatrist. The profile of sexual murderers he developed was that of a male who was younger than age 35 years, introverted, a loner, and with a rich fantasy life. A close, but ambivalent, tie with his mother existed in which she was seen as someone inordinately curious about his sexual life. A distant and punitive father produced a difficult relationship between the father and son. The future murderer was seen as vain and sensitive to threats to his self-esteem as well as having concerns about his sexual potency. Accounts of cruelties appealed to him, yet a mixture of kind acts would be recorded about him as well. A homicide could occur in the course of a sexual assault, such as by strangulation. In some cases, sexual intercourse itself did not take place; masturbation might occur, or some bizarre type of mutilation, dismemberment, or stabbing. The men Brittain studied were seen as having an intellectual grasp of the crime but showing only a flat and superficial emotional response to it. Whereas some of the preplanning to their acts suggested cunning, they often returned to the scene of the crime and became upset if there was a lack of publicity. West9 emphasized the difficulty in generalizing about sexual homicides, given that there are often so many unique and distracting features to each case. Some sexual killers have a pattern in which aggression periodically breaks through, whereas others have impulses that remain on a fantasy level and then only break out in one violent and irrational act. In these cases, the usual defenses used against guilt and hostility have given way. Once the barrier is broken, the behavior seems more likely to recur and possibly become more extreme. However, this pattern would correspond to cases that have long intervals between sexual homicides, in which the perpetrator remained dormant. Sexual killers may share many characteristics with those who commit sexually sadistic acts, in which violence is mixed with a need to dominate and degrade another. As noted, a recurrent and key question is what elements shift the person from fantasy to actual enactments of the sadistic sexual fantasies. A second question that applies to those who act on their fantasies is what elements shift the individuals further into becoming sexual murderers. Between the fantasies and commission of murder are a host of fantasies and activities in which sadistic behavior may occur in a legal or

342


illegal context. Sexually sadistic fantasies that involve a sexually arousing component are believed to predispose to sexual homicides, in the sense of lust homicides. However, the key in the fantasies, and in a final homicidal enactment, is the sense of mastery and power over another. The paramount feature is the persistent pattern of sexual arousal connected with images of pain and suffering inflicted on someone under the individual’s control. However, the fantasies of pain and suffering need not have been present in their final form for years but could have evolved more recently. Dietz and colleagues’10 study of 30 sexually sadistic criminals in the files of the National Center for the Analysis of Violent Crime revealed that 22 of the subjects were known to have murdered. In fact, there were 187 murder victims, with five men responsible for 122 of the murders. Seventeen of the men were suspected of being serial killers. The ultimate cause of death could be established for 130 victims: 61% died by asphyxiation, whether through ligature manual strangulation, hanging, or suffocation; and only 25% were killed by a gunshot wound. In the serial killings, the different victims were killed usually by inconsistent methods. However, even though only 3% of the victims died from a beating, 60% had been beaten before death. Fifty-seven percent of the perpetrators victimized only adults, whereas 43% victimized children. Sexually sadistic individuals have some peculiar characteristics that accompany their offenses. Details about individual fantasies involving torture or arousing sadistic fantasies are important in trying to connect these thoughts with the act of homicide. Some of these individuals engaged in what was referred to by Dietz’s group as excessive driving, in which they would drive long distances with a lack of any clear goal and then abruptly change direction or circle back. This behavior seemed to portray their need to feel free to do as they wished along with the compulsive need to dominate and control others. These individuals exhibited a type of pathological development in the quest for autonomy. Their fascination with police activities and paraphernalia similarly corresponded to their power needs and identifications. Before being killed, the victims may have been held captive for varying lengths of time to allow the perpetrator to exert further control on them or to inflict more pain and suffering. Victims may have been told to utter certain phrases to correspond to the killer’s fantasies. Personal belongings of the victims might have been kept as trophies or souvenirs to signify conquests or to use at other times in recreating states of sexual arousal. Jeffrey Dahmer collected the skulls of his victims as well as pictures of them and later of their corpses. This behavior is congruent with Dietz and colleagues’ study,10 in which 53% of the sexually sadistic offenders retained records of their crimes, such as drawings, photographs,

Sexual Homicide

343

electronic recordings, or the victim’s bones. Such practices suggest an unemotional, detached type of perpetrator who lacks empathy with other human beings. The narcissistic absorption would be seen as part of the psychopathy of the perpetrators. A not infrequent phrase heard in their descriptions of the sexual homicides was “to get on with the business.”

Case of David Berkowitz The profile of one well-publicized sexual killer—Son of Sam—may be instructive at this point. Son of Sam was an individual who left notes signed with this appellation, seemingly either to challenge or baffle the police. Abrahamson,11 a New York psychiatrist, had the opportunity to investigate this killer over 4 years. The case raised many issues that are still under discussion. Although eight young women were shot and killed in 1 year’s time, none were sexually attacked as such. A sexual motive to the killings was hypothesized from the context of these shootings, in which many of the victims were in parked cars with their boyfriends at night. Some of the males were shot and killed as well. It was clear that these were a serial type of killing, but it could not be automatically inferred that these were necessarily sexual homicides as distinguished from the killer’s acting out of sexualized fantasies and conflicts. The killer was arrested shortly after the eighth homicide and was revealed to be a 24-yearold male named David Berkowitz. He had given himself away by writing threatening letters to those who annoyed him. One note was to a neighbor whose barking dog upset him; he eventually killed the dog. Another letter to a neighbor referred to demons and the threat that he would light a fire outside the neighbor’s door. Setting fires in old buildings, cars, or rubbish heaps appealed to Berkowitz, and such arsonous acts sometimes preceded the murders. He kept a diary in which he recorded 1,488 fires he had set, along with torturing animals.12 These acts can be seen as reflections of his need to have control over events along the pathway to homicides. Berkowitz revealed a troubled childhood in which he had been adopted. Throughout his childhood he remained close to his adoptive mother, who indulged him, but he was resented by his father. When he was 15 years old, a major trauma occurred when his mother died of cancer. At that time, he became depressed. When the father remarried, there was a further estrangement between them. While in the U.S. Army, Berkowitz got into trouble with the authority structure. He later tried joining a fundamentalist religious group. His sexual views were strict, and he saw sex outside marriage as a serious sin; however, he masturbated several times daily and remained shy of women.

344


A search for his biological parents upset Berkowitz even more. His mother was revealed as having been the mistress of a married man who later died, but most disturbing to him was that his mother had kept a daughter, although she had put him up for adoption. He described experiencing feelings of rejection and humiliation following this news, both of which were seen as the basis for a revenge type of homicide. Berkowitz’s first attempt at homicide, when he stabbed a woman, failed. Later he planned to carry out the killings with guns. The process of seeking out young women, as if he were involved in a hunt, gave him an emotional high. After a shooting, he would return to his seemingly normal pattern of existence, such as going to work the next morning. At his trial, the psychiatric opinions differed between those clinicians who saw him as psychotic and driven by demons to commit violent acts versus those clinicians who saw the demons simply as a way of his expressing a bad side of himself. He eventually pled guilty to the killings and was sentenced to 547 years in prison.

DISMEMBERMENT CASES Sexual homicides need not be heterosexual in nature.

Case of Dennis Nilsen Dennis Nilsen was a homosexual government employee in London who killed at least 15 young males over a 4-year period ending in 1983.13 The victims were mainly boys he would pick up in bars or on the street and who were under the influence of alcohol. He would offer them a place to sleep and subsequently strangle them. He would then dismember their bodies and dispose of them by putting some parts in garbage bins, burying other parts in gardens, and burning other parts. Later he boiled his victims’ flesh and flushed it down a toilet. This practice eventually blocked the sewage system; sewer workers eventually found the human flesh. Based on his writings found after he was arrested, in which he described how attractive he found the bodies of dead young men, such as “the limpness of the movable parts…the texture of dead, cold skin to the touch,” it was clear he had necrophilic interests.14 There is a striking similarity in the Nilsen case to one that I had the opportunity to evaluate in a legal setting after a man had been charged with the murder of his wife.15 He had strangled her and dismembered her body and had begun to flush parts of her body down the garbage disposal. When it eventually became plugged, he had to stop. All of the dismember-

Sexual Homicide

345

ment was carried out after the strangulation; he made trips from the bedroom where the killing had occurred to the kitchen where the garbage disposal was located. His mother, who lived with the couple, sat in the living room throughout all of these proceedings and said nothing. She was later charged as an accessory to the crime and convicted. He denied any previous sexual assaults or dismemberments.

Case of Jeffrey Dahmer The case of Jeffrey Dahmer in Milwaukee, WI, fits many of these behaviors as well.16 Dahmer was charged with 13 counts of first-degree intentional homicide and 2 counts of first-degree murder after he confessed to killing 17 men. He was reported as having made the confession in a spirit of remorse and not in any spirit of bravado or satisfaction. Dahmer would kill his victim, undress and caress the dead body that he now possessed, and masturbate. He would then dispose of the corpses very quickly by using an electric saw and acid baths. However, he boiled the heads and saved several of them. The jury had trouble with the fact that Dahmer did not seem to enjoy torturing his victims while they were alive but obtained pleasure only when they were dead, a form of necrophilia. Six years before his arrest for the homicides, Dahmer had been charged with disorderly conduct for exposing himself to children. He was found guilty and placed on 1-year probation. Two years later he picked up a Laotian boy to pose for photographs, gave him a drink laced with a sleeping potion, and then fondled him. At the time, he was charged with second-degree sexual assault and enticement of a child for immoral purposes and sentenced to 8 years in prison, but after he expressed contrition, the sentence was reduced to 1 year’s detention with 5 years’ probation. He was then to receive psychological treatment to deal with his sexual confusion and alcohol dependence. Neighbors had seen Dahmer as a well-mannered, polite boy who kept to himself. At age 4 years, he had had a double hernia operation and suffered great pain, which he remembered all his life as giving him the feeling that his genitals had been cut off. A high school teacher saw him as solemn and depressed. As a boy, he would kill squirrels and rabbits, skin and boil them, and preserve their heads for secret rites.17 Colleagues at work and those who knew him in the army described how he would become a dramatically changed, aggressive man when he became intoxicated. A 6.5year period existed between his first homicide and the second, but thereafter he began the pattern of serial sexual homicides. Again, a significant clinical question is what permits these patterns to stay in check over many years, and what then promotes the beginning of a compulsive pattern of

346


homicide. Some see the public’s obsession about cases like Dahmer as saying as much about our society as about the killers themselves. Dahmer has been viewed as part of our entertainment culture, a figure of impurity who needs to be cleansed from our society. In this regard, note that besides the president and pope, Dahmer was one of the few people most Americans could identify.18 He was murdered in prison after his conviction.

CHILDREN AS VICTIMS OF SEXUAL HOMICIDE Child sexual homicides may be of a heterosexual nature, homosexual nature, or a mixture of both. Again, the same types of homicides exist as with adult victims: a sexual act may have taken place and the homicide could have occurred in the context of the sexual act; the homicide could have been part of a planned act of lust homicide; or the homicide may have taken place to prevent a later discovery of the sexual acts themselves. A distinction should be made between victims of a prepubertal age versus adolescents. Young children are chosen because they are seen as more vulnerable to being lured away and are therefore more helpless. Correspondingly, children are more vulnerable to enabling the perpetrator to play out themes of dominance. Adolescent victims may come from a troubled family background and wander around the country, perhaps becoming caught up in a shiftless lifestyle in major metropolitan areas. Although some of these adolescents are reported missing, not all are. In contrast, young children are almost always reported missing, but some are never heard from again and may become victims of this type of sexual homicide. The moors murders in England achieved great publicity at the time when Ian Brady and Myra Hindley, his female lover, tortured and killed several children and then buried them on the moors.19 The perpetrators were eventually arrested and tried for the murders of a girl age 10 years and two boys ages 12 and 17 years, although it was felt that they may have been responsible for the disappearance of various other missing children. They tape-recorded the murder of a 10-year-old; the child’s agony and pleas could be heard. The slavish devotion of Hindley as an accomplice and accommodator was striking.20 Patterns cannot always be assumed to be fixed in nature, as is seen in the following case example.

Case Example A 19-year-old male college student was charged with the death of an 8year-old girl. He had first lured her into the basement of the store where

Sexual Homicide

347

he worked, stunned her by hitting her with a hammer, and then placed her in a box and driven to his parents’ home, where he lived in the basement. A day later he took the girl, still in the box, to his brother’s apartment, which he was watching while his brother was on vacation. There he began to sexually explore the girl with his finger, including her vaginal and anal areas, while in the shower with her. He paid great attention to her cleanliness. His explanation for the killing was that it occurred in the context of debating within himself whether he should take her to a police station and tell the police what happened versus trying to hide the evidence. When he opted for the latter, it meant she would have to die. He filled the bathtub with water and lay on top of the girl until she drowned. He reported experiencing no sexual excitement while killing her. In this case, the perpetrator had developed a fantasy of holding someone captive 9 months earlier, and for 6 months the fantasy had involved holding an adult woman captive. The fantasy then began slowly to shift to a young girl. Various combinations of fantasies were used and would recur before the final one was chosen. He described experiencing shifting personality states with different accompanying moods. One was an angry and irritable self that came to the fore when his fantasies were of abduction in contrast to his normal self, which was seen as a pleasant person and manifest when he went to school or work. In the past, he had told male companions that he had thoughts of killing people, but they saw him as only joking.

CHILDREN AS PERPETRATORS OF SEXUAL HOMICIDE The topic of children committing homicide has acquired a good deal of national publicity, particularly for those children who are not yet into adolescence. However, these cases are rarely sexual homicides, although some have had a sexual connection. The majority of cases are adolescents who have had thoughts and fantasies involving sexual acts and violence that were then acted out. Myers21 studied 16 male adolescents who had committed a sexual homicide and classified them into four types. An explosive type had a sudden release of sexual and aggressive feelings; a predatory type sought out victims; a revenge type acted out unresolved anger toward someone the adolescent felt had wronged him; and a displaced matricide type was an adolescent who displaced his anger toward his mother to another female. The latter was the least common. Almost all of these adolescents carried the ubiquitous diagnosis of conduct disorder. However, some intriguing findings are related to neuropsychiatric vulnerabilities, impairments in the capacity to feel guilt, and school problems. Axis II diagnoses were usually schizoid or schizotypal personality disorders with some transient psychotic symptoms.

348


PSYCHODYNAMICS OF SEXUAL MURDERERS Almost everyone who has studied or assessed sexual murderers is impressed by the role that fantasy has played in the murderers’ lives with respect to the eventual killing. One perspective is to see these murderers as basically exhibiting the hallmarks of a perversion or paraphilia, leading to a lust homicide. Many of the primitive and polymorphous fantasies of the perpetrator as a young child are fused into the murderous act. Although the resulting sexual homicide may not be an enactment of the preexisting fantasies, killing usually occurs because the urge to act on the fantasies in some form eventually breaks through the perpetrator’s ego defenses, with diverse outcomes.

Use of Denial A mixture of tender behavior with brutal imagery may be present in sexual killers, with the balance tipping from one extreme to the other from time to time. The use of denial is often mentioned in connection with sexual murderers, as it is with sexual offenders in general. Caution is needed here because denial is often used on different levels of explanation.22 For example, when denial is conscious, it refers to an individual simply trying to deny the acts outright. Sometimes denial is used to mean taking a position that the behavior was something justified, such as the perpetrator being provoked into the act or feeling that he or she simply became “carried away” and went a bit too far. Denial may also be used as a form of excuse, such as being intoxicated or on drugs at the time the attack occurred. These types of explanations, in which denial is prominent, are quite different from the type of unconscious denial mechanism that is present in the sexual murderer. The unconscious denial in this group is more in the nature of a splitting mechanism, in which one part of the perpetrator becomes horrified at the gruesome act that has been committed. Some experts have conceptualized this response as a subsequent acting out of past childhood traumatic situations in which a screen memory has taken the place of actual events. The idea of various types of traumatic acts from childhood emerging in a distorted form in the adult is congruent with current theories about the effects of childhood abuse and deprivation.23 The unresolved question is why such earlier abuse leads to an outcome of a sexual homicide in only a few individuals when child abuse appears so much more prevalent and can have diverse outcomes. Perhaps if there were opportunities to study and treat these individuals over a prolonged period of time, more answers would be available.

Sexual Homicide

349

Role of Fantasies What has been confirmed by clinical observations on sexual killers is that they are not usually outright psychotic. They are more likely individuals who periodically become obsessed with their fantasies. In contrast to the majority of individuals who have such fantasies, something promotes these particular people to literally act out their preoccupation, although often with embellishments. At times other than at the enactment of a brutal homicide, they lead what appears to be a normal life, engaging in work and other daily living activities, so that no one else notices anything unusual. It is also significant that the fantasies may not be acted on for years, although, in retrospect, if one had the opportunity to work with these individuals before a homicide, indications of the types of instabilities seen in certain personality disturbances would probably have been present. A question is then whether perverse fantasies have been present and become expansive over time in those with certain types of personality disorders, such as borderline or narcissistic. When the sexual homicide takes place, it is seen as an enactment in the external world of the internal situation that has been brooded about in the unconscious for some time.24 When these fantasies break out by mobilizing themselves into action in the external world, it can be something like an explosion or what was described several decades ago by Wertham25 as a “catathymic crisis” or part of an inner driven compulsion that gets unleashed periodically.

Splitting The part of the person that carries out the torture, mutilation, and killing is not completely repressed in the sense of a dissociated state. The state can more accurately be described as a splitting of the personality, in which one part of the personality is in the ascendancy, such as the part that carries out the acts, while the other, acquiescing part observes the events. However, the joint participation of these selves allows the events to be recalled. The perpetrating part of the self that carries out cruel and sadistic acts of torture may ultimately be seen as the good part of the self, carrying out these acts on another who stands for a projected, evil part of the self. Hence, Jack the Ripper selected prostitutes as victims, a not infrequent theme found in certain sexual murderers. Another possibility is that the evil or bad part of one or more of these others is introjected so that this part is in control of the self. Quite rapid alternations can occur between these parts, which explain puzzling acts of kindness or tenderness that get mixed in with acts of brutality and

350


eventually homicide. This could explain such puzzling phenomena as the sexual killer who takes great pains to make a victim comfortable before he tortures him or her or even in the act of torture wants the victim comfortable on pillows, such as in the following case example.

Case Example A young woman was kidnapped and taken to an isolated area of a park during the middle of the night. She was tied down and beaten with fists over the course of several hours. At one point, she was told that she was to be offered as a sacrifice. The perpetrator then began cutting off her lower leg with a knife in an effort to remove her ankle. Partway through the amputation, before any permanent damage had been sustained, he stopped in the middle of her screams and began to feel sorry for her. He proceeded to untie her and wrapped her in a blanket, expressing concern that she might be cold, and drove her to the hospital emergency department. After carrying her into the hospital and placing her in a wheelchair, he walked out. An alert nurse, sensing something was strange, copied down his license plate, which led to his later apprehension.

This case is similar to that of Jerome Brudos, who had had a fetish about women’s shoes since childhood.26 As an adult he killed four women, cut off their feet and breasts, and dressed them in his collection of women’s clothing. The amputated feet were used to display his collection of women’s shoes. A variation of the splitting process occurs when certain aspects of the self are disowned or disavowed. An image of the potential victim to be selected emerges in the fantasy, which may initially be vague. For example, the perpetrator may first focus on the image of a young woman, then shift to the image of an older woman. In other cases, the victim eventually selected is unknown and emerges on some chance basis of convenience, even though the fantasy and preoccupation with selecting such a victim may have been present for months or years. On occasion, the full fantasy has not been played out beforehand. For example, a child may be selected to be the projected depository for the hate, envy, and other feelings being experienced within the perpetrator; thus, later acts of torture or killing that child allow the perpetrator to permanently sever the bad aspects of the self from the individual. Or in the case of the 19-year-old man who kidnapped the 8-year-old girl (in the earlier section “Children as Victims of Sexual Homicide”), the perpetrator’s fantasy centered on knocking a girl unconscious by hitting her beforehand, holding her captive for an indefinite time, and engaging in sexual acts with her. The fantasy in-

Sexual Homicide

351

volved thoughts of treating her kindly when she was held as a captive, perhaps as a kind parent would, such as giving her a better home, food, and clothing than her own family had provided. At one point, the perpetrator tried out the fantasy. It is this constant potential to shift out of this kind, nurturing role and to allow another aspect of the self to become dominant that makes these situations so precarious. That is one reason that prevention and prediction are made difficult. One of the disturbing things about these fantasies and enactments is that they serve complicated defensive and adaptive functions in which diverse types of psychopathology are present. Sometimes the fantasies may serve the function of binding aggression and keeping the person integrated.27 Dual aspects of cruel treatment mingled with compassionate concern are not confined to sexual homicides. They simply reflect the process of splitting. Acts of cruelty to animals, including killing, may be enacted in such a framework. If a pet is killed or tortured, a severe depression may later emerge accompanied by belated attempts at expiation. However, alternations between compassion and cruelty are significant clinical findings when working with any person who has paraphiliac problems. In such persons, the potential for assigning another person the good or bad aspects of the self via a projective identification mechanism holds the possibility for a miscarriage in either direction. The good self may be as likely a candidate for riddance as the bad self. Similarly, both homicidal and suicidal impulses may be simultaneously present. In the following case example, the interplay between the good and bad self is seen.

Case Example While holding a woman captive who was tied in a chair, the perpetrator alternated between sexually molesting her and washing her feet while crying along with her. He later recalled he was crying about her upcoming death, but at the same time he did not want to hurt her. Because there was no way to kill her without hurting her, he could only undo the upcoming acts of molesting and killing up to a certain point by acting as her servant and joining her in crying. When these “reparations” ceased to be successful, a crisis point was reached and a homicidal act occurred.

Case Example Dennis Rader was a respected member of his community, a Cub Scout leader, and president of his church council. He was also a serial sexual murderer. Married with two children, he lived in a suburb of Wichita, KS. He had labeled himself as the “B.T.K.” killer, which stood for “bind, torture, and kill”—the processes he used in carrying out his killings. His first victims in 1974 were members of a family he had stalked and later

352


killed, including the parents, a 11-year-old daughter, and her 9-year-old brother. The girl was first hung from a sewer pipe in the basement, with her panties yanked down over her bound ankles, as he began to strangle her. She was squealing, “Momma, Momma,” and asked what would become of her. His reply was, “Well, honey, you’re going to be in heaven tonight with the rest of your family.” He had let the girl’s toes dangle a fraction of an inch from the floor to slow her death while he masturbated. Subsequent killings went on over a 30-year period in which he stalked hundreds of potential victims and compiled volumes of notes on them; he referred to them as his “projects” and kept the notes in a green box he called his “mother lode.” One victim was strangled and then taken to a church and left in a bondage position. He would strangle people to the point of them passing out, then loosen his grip until they regained consciousness and repeat the strangling. These acts amounted to committing multiple murders with the same victim before finally killing him or her.28 He is currently appealing a life sentence in the Kansas judicial system for 10 murders.

CATATHYMIC AND COMPULSIVE SEXUAL HOMICIDE Schlesinger29 elaborated on a psychodynamic or motivational scheme of sexual homicides between two overall types—catathymic and compulsive. The categories are not seen as diagnoses and the two types can either be planned or unplanned. The catathymic idea is an elaboration on some of the original work of the Swiss psychiatrist Hans Maier in the early twentieth century. The concept was later taken up by Fredric Wertham, who proposed the concept of the “catathymic crisis” in his work on violence and homicide. Acute catathymic homicides as now formulated by Schlesinger are seen as a sudden loss of control resulting in a sexual homicide caused by a breakthrough of underlying sexual conflicts. The acute type is triggered by some external circumstance. Such explosive and sudden murderous assaults were seen as precipitated by some type of threat or perceived insult that involved the adequacy of the person in a relationship and the stability of the relationship. Examples of precipitating factors are a belittling rejection by a sexually provocative paramour, teasing or hostile remarks, or a sadistic threat by a sexual partner. Chronic catathymic homicides have an incubation period in which the person becomes depressed and obsessively preoccupied with the future victim. The idea to kill emerges slowly and is often communicated to friends, family, or professionals but ignored when the person assures them that he or she is in control. In time the idea to kill gains in strength, takes hold, and becomes fixed and intractable. A homicide or some other

Sexual Homicide

353

serious act of violence may occur, following which the person simultaneously has a sense of relief yet horror at what he or she has done. A literary example of this type of killing is Dostoyevsky’s Crime and Punishment. Meloy30 believes that those who commit a chronic catathymic homicide have at least a borderline or psychotic personality organization. The formulation is that malevolent and controlling characteristics are attributed to the future victim and that the perpetrator’s state of mind stems from a disorder of attachment. The explosive outburst is viewed as primarily affective in nature, although it can be predatory if planning has been present. The distinction is that it is not simply a depressed state per se that results in a sexual homicide but the emergence of the idea to kill accompanied by pressure to do the act—the catathymic motivational process. Compulsive sexual homicide is seen as a fusion of sex and aggression wherein a powerful internal drive leads to seeking out victims to kill, which is sexually gratifying. When unplanned, the urge breaks through the person’s controls when the opportunity presents itself or a victim is selected. This type of killing seems to describe the traditional type of lust homicide. Violent thoughts and fantasies may go on for years before an eruption. Again, the challenging question is what keeps the impulse to kill in check over time until it is finally acted on at a particular moment. These compulsive offenders have aberrant sexual fantasies in which gratification is achieved through various forms of aggression. The gratification may thus be sought in repetitive behavior. Schlesinger29 noted that these individuals are portrayed as having, and claim to have, an irresistible impulse, but they usually choose not to exercise restraint because they seek relief from an inner tension. This inner tension is contrasted with the change in thinking associated with sexual conflicts in catathymic killings in which there is also a triggering challenge. Although there are many examples of the compulsive sexual murderer throughout history and in different countries, two well-known cases from the United States can provide illustration.

Case Examples William Heirens was a University of Chicago student with a fetish for female underwear, which led him to commit several burglaries. He had killed two adult women in their homes and then abducted and killed a 6-year-old girl and deposited parts of her body in sewers. When he tried to resist the inner compulsion to kill, he developed headaches and profuse sweating. In the home of one of his victims, he wrote on a mirror in lipstick, “For Heaven sakes catch me before I kill more, I cannot control myself.”31

354


Edward Gein, a Wisconsin farmer, was the actual model for the Hitchcock film Psycho.32 After his mother’s death, he preserved a room for her as a shrine. During her life, she had conveyed to him ideas about the sinful nature of women. Gein began to dig up corpses from cemeteries and then to kill and disembowel women at his home. Police discovered a victim’s headless corpse hanging upside down in his home, chairs upholstered with human skin, and a vest Gein had made for himself out of human skin.

SUMMARY OF APPLICABLE PSYCHODYNAMIC HYPOTHESES Innate aggressive tendencies can fuse with those that are sexual and may vary biologically from person to person. Apart from this perspective, some psychodynamic hypotheses can be offered about the sexual murderer based on current knowledge. The following are summary points of these hypotheses. 1. In most cases, the victim is an unknown or anonymous person who is selected because of that fact. 2. The antecedents of a sexual homicide are present in quiescent form for a long period of time before a homicide occurs. 3. What finally precipitates a homicide is difficult to determine because homicides do not arise in the majority of those with sadistically perverse fantasies. In some cases, the incubated fantasy is primed and ready to be acted on whenever the right circumstances arise. In some cases, stalking behavior might have been involved in the fantasy, with multiple individuals being stalked who are unaware of being stalked or of what might someday await them. Again, the prediction problem is present in not having the ability to predict whether anything beyond stalking may ever occur. The research of Dietz et al.33,34 on letters to celebrities and congressional representatives goes to the question of which individuals may be approached, but gives no predictive basis as to whether some type of attack might ever occur. 4. Unresolved mixtures of love and hate exist in an amorphous state at different times in sexual killers; these mixtures are more characteristic of the mental life of a young child. 5. Themes of revenge and envy may become prominent as a danger period of action is approached. 6. The victim selected is often an innocent scapegoat for these tempestuous and unresolved feelings. 7. Acts of kindness and reparation may be carried out simultaneously with those of the utmost cruelty.

Sexual Homicide

355

8. A displacement onto the victim of frightening internal images takes place that seems to initiate destructive impulses. 9. Although sadomasochism is an essential part of normal sexual functioning, it can also present as a perversion. For those who attempt or commit a sexual homicide, the behavior extends beyond the restricted perverse activities of simply deriving pleasure from inflicting or receiving pain as a condition for sexual gratification. A variety of different clinical types may be present in which aggressive components are acted out in the context of regression or disintegration of the superego, splitting, pathological grandiosity, and a loss of ego boundaries.35 Themes of domination, and the need to have a victim under one’s control, are the driving forces. 10. Fetishistic objects may be used when ritualistic aspects accompany a sexual homicide. The popular media often portray these as cult or devil worship events, but they are more likely related to torturing the victims before death by use of instruments, electricity, beatings, or burnings or by tying up the victim in painful positions. A ritualistic “signature” may be left at the site. 11. In some cases, a childhood history of overt physical or sexual abuse may be present, but the lack of specificity for this finding mitigates it as a generalization. It is also possible to find terrifying images of authority figures that have been internalized from overindulgent parental figures. 12. Distinctions between catathymic types of sexual homicides and the compulsive variety that have a drive to repeat the homicide due to its gratifying aspects can give specificity to some killings. 13. Finally, lacking knowledge about the specific internal sexual and aggressive conflicts of a person, as well as the specific triggers that exist for certain individuals, makes prediction and prevention a difficult task. Without a resolution of these types of internalized conflicts and fantasies, serious questions remain about whether sexual killers will feel the need to repeat their violent behavior.

LEGISLATIVE AND JUDICIAL RESPONSES Sexual violence on a homicidal level is a powerful stimulus for governmental and political groups to take some action. Not only is sexual violence a loaded issue because it involves sexual acts often perpetrated against women or children, but it also influences the personalities of those in power to try to do something in response to the violence. State laws have attempted to deal with the more serious sexual offenders.

356


Such measures have always had a dual, if not competing, theme: treatment and rehabilitation of the offender versus concern about the dangerousness and recidivism for community protection. Implementing such measures has always varied depending on the particular historical period. Although these dual goals have also existed for how society has dealt with its mentally ill, the matter becomes more complicated when dealing with sexual offenders. A major difference between the mentally ill and most of the diverse types of sexual offenders, including those who commit homicides, is that the latter are not psychotic, at least by descriptive criteria that note symptoms of delusions and hallucinations. The difference between these two groups is problematic because legislatures and courts may employ the same words or classifications used in medicine and psychiatry but use their own definitions and interpretations. Legal confusion results when psychiatrists testify on these matters from the perspective of their own profession. Much debate and argument at the trial court level, and on appeal, may center on issues about the degree of control a person had over his or her violent sexual behavior. Philosophers have not resolved free will issues over the past 2,000 years, and the matter becomes doubly confusing when lawyers and psychiatrists enter the debate and, in an adversarial context, argue the pros and cons about the issue of control over sexual behavior. Other diverse legal, clinical, and societal issues surround acts of sexual violence and those who commit these offenses; these issues are beyond the scope of this book. These questions involve the validity of clinical diagnoses for those charged with a sexual offense, how psychiatrists may testify when the classification used is a legislative one, types of treatments available and their success or failure rate, juveniles who commit sexual homicides or become victims of sexual attacks, and the recurrent problem of risk assessments. Historically, the modern era of statutes dealing with sexual offenders, beyond the direct criminal conviction as for any other crime, began in the 1930s. A “psychopathic personality” statute passed in Minnesota was the first to be upheld as constitutional by the U.S. Supreme Court in 1940.36 As an illustration of future problems, the legislature gave its own definition of a psychopathic personality as a person with “such conditions of emotional instability or impulsiveness of behavior, or lack of customary standards of good judgment, or failure to appreciate the consequences of his acts, or a combination of such conditions, as to render such a person irresponsible for his conduct with respect to sexual matters and thereby dangerous to other persons.”37 The interpretation of the Minnesota Supreme Court, upheld by the U.S. Supreme Court,

Sexual Homicide

357

was that this provision meant that these persons had an “utter lack of power to control their sexual impulses” and referred to their “uncontrolled and uncontrollable desire.” This ruling opened the door for many states to pass similar legislation. Some states required an initial criminal conviction, whereas others allowed a civil commitment under the statute without any criminal conviction. Upon a commitment, some states would send the person to a hospital and then later require any remaining time of a sentence to be served in prison. Over time the statutes were used less frequently or repealed. By 1990, only 13 of the 28 states that had passed sexual psychopath laws still retained them.38 The Group for the Advancement of Psychiatry39 described the sexual psychopath statutes as an approach that had failed and recommended that they therefore be halted. A question is what led to the resurrection of these statutes and their present more frequent and aggressive use. The failure to rehabilitate and prevent recidivism of many sexual offenders was one factor that had emerged by the 1990s. Increased lengths of sentences upon conviction and determinate sentences had replaced the indeterminate ranges for time imprisoned. Ironically, determinate sentences resulted in the release of some sexual offenders who then repeated their violent sexual behavior. States then sought some way to continue detaining individuals who had served their prison time after a conviction and were ready for release.40 Because determinate sentencing was found insufficient, a reversion to additional civil commitment measures for sexual offenders was resuscitated. States then either refurbished their old statutes or created new commitment laws, with such titles as sexually dangerous persons or sexually violent predators (SVPs). Committing individuals as SVPs allowed their continued confinement under the rubric of a civil commitment. Although most of the sexual offenders were not mentally ill, legislatures used words that sounded psychiatric but actually had their own legislative meaning or were left undefined. Typical legislation called for proof of a “mental abnormality” or a “personality disorder” that would meet the criteria for a sexual predator. Eight states defined mental abnormality as “a congenital or acquired condition affecting the emotional or volitional capacity that predisposes the person to commit sexually violent acts” (p. 356).41 Many problems arose, such as antisocial personality disorder being viewed as a mental abnormality. If that view was adopted, the question was how sexual predators were any different from those simply convicted of a sexual offense and sent to prison or the large numbers with antisocial personality imprisoned for diverse offenses.

358


These issues culminated in a challenge to an SVP statute in the case of Kansas v. Crane. The case was a challenge under due process on the criteria for commitment based on a “mental abnormality or personality disorder.”42 Crane was a convicted sex offender with diagnoses of antisocial personality disorder and exhibitionism. After his conviction, he was committed as an SVP, but the Kansas Supreme Court reversed the conviction, holding that a previous case decided by the U.S. Supreme Court in 1997 required that the person not be able to control his or her behavior for such a commitment. The intricacies of the case reveal the reasoning. First, a finding of dangerousness had to be tied to the existence of a “mental abnormality” or “personality disorder” that made it “difficult if not impossible” to control the dangerous behavior. The case was then appealed to the Supreme Court to decide whether the state always had to prove that a dangerous individual was completely unable to control his or her behavior. In reversing the Kansas Supreme Court, the ruling was that there was not a requirement for an absolute lack of control and it would suffice if the “abnormality” simply created a “serious difficulty” in control. Psychiatry could perhaps play a role in providing some clinical information about the individual to be committed as an SVP, but the matter was always a legal determination. Risk assessment of sexual offenders is a persistent problem. Assessing the risk for dangerousness may begin at the point of an initial arrest for a sexual offense, after a conviction, at the time of a rearrest when on probation, or at the point of release from a sentence served when a petition for a civil commitment as an SVP is filed. Parallel to the expansion of sexual predator statutes, various testing instruments have evolved to try to assess the risk of recidivism.43 For diverse reasons, there are major obstacles in their validity and reliability, especially if they are used to try to predict future homicidal sexual assaults. Studies vary in definitions of terms, the groups used for study, attempts to compare groups or have some type of control group, and the depth of data available about the subjects. Hanson44 pointed out that the accuracy of expert opinions in predicting sexual recidivism is only slightly above chance levels, with an average r of 0.10. Most sexual offenders are not recidivists. Hence, some mechanism or instrument is needed that would focus on a small group of high-risk sex offenders whose probability of recidivism is greater. Many of the test instruments focus on “static” factors, such as historical data regarding past deviant sexual practices, prior sexual offenses, and early onset of sex offending. In contrast, use of “dynamic” factors improves the accuracy of risk assessments.45 Thus when recidivists are compared with nonrecidivists, the former have intimacy deficits seen as

Sexual Homicide

359

problems in forming long-term love relationships, poor social networking, and inadequate coping strategies in which they resort to sexual fantasies or behavior to cope with stressful life events. In summary, there is a long way to go in understanding the genesis of violent sexual behavior that may end in a homicide. Multiple forms of such behavior require specificity for understanding if adequate preventive measures are ever to be achieved. Meanwhile, what has evolved is an amalgam of law, clinical appraisals, societal fears, attempts at creating assessment instruments, and a confusing set of statutes used to detain individuals beyond the time served for a criminal conviction. A prosecutorial view is that these types of statutes provide “one arrow in the quiver” for preventing violent sexual offenses.46 Tonry47 pointed out that few judges or informed scholars support such laws in the forms in which they have been adopted because they are too rigid and often result in unjustly harsh penalties. This outcome leads to efforts to circumvent these laws by judges and lawyers in individual cases. Although these statutes may achieve a goal of detaining some individuals for long periods of time on the basis of public safety and thereby prevent some sexual homicides, they achieve little else in the way of prevention.48 The processes of determining who is committed after serving a prison sentence already are quite unpredictable and create a sense of injustice. At present, longer prison terms have seemed to provide a temporary solution. In the long run, much better clinical and actuarial data will hopefully offer better community protection along with a sense of fairness.

REFERENCES 1. Krafft-Ebing RV: Psychopathia Sexualis (1898). Translated by Wedeck H. New York, Putnam, 1965 2. Ressler RK, Burgess AW, Douglas JE: Sexual Homicide: Patterns and Motives. Lexington, KY, Lexington Books, 1988 3. Glasser M, Kolvin I, Campbell D, et al: Cycle of child sexual abuse: links between being a victim and becoming a perpetrator. Br J Psychiatry 179:482– 494, 2001 4. Douglas JE, Burgess AW, Burgess AG, et al: Crime Classification Manual: A Standard System for Investigating and Classifying Violent Crime. New York, Simon & Schuster, 1992 5. Canter DV, Alison LJ, Alison E, et al: The organized/disorganized typology of serial murder: myth or model? Psychol Public Policy Law 10:293–320, 2004 6. Rule A: Green River, Running Red: The Real Story of the Green River Killer, America’s Deadliest Serial Murderer. New York, Free Press, 2004

360


7. Gibson E, Klein S: Murder 1957–1968 (Home Office Research Studies No. 3). London, England, HM Stationary Office, 1969 8. Brittain RP: The sadistic murderer. Med Sci Law 10:198–207, 1970 9. West DJ: Sexual Crimes and Confrontations. Brookfield, VT, Gower Publishing, 1987 10. Dietz PE, Hazelwood RR, Warren J: The sexually sadistic criminal and his offenses. Bull Am Acad Psychiatry Law 18:163–178, 1990 11. Abrahamson D: Confessions of Son of Sam. New York, Columbia University Press, 1985 12. Ressler RK, Schachtman T: Whoever Fights Monsters. New York, St. Martin’s Press, 1992 13. McConnell B, Bence D: The Nilsen File. London, England, MacDonaldFuture, 1983 14. Masters B: Killing for Company: The Case of Dennis Nilsen. London, England, Jonathan Cope, 1985 15. State v Hoffman, 328 NW2d 709 (1982) 16. Masters B: The Shrine of Jeffrey Dahmer. London, England, Hodder and Stoughton, 1993 17. Norris J: Jeffrey Dahmer. London, England, Constable, 1993 18. Seltzer M: Serial Killers: Death and Life in America’s World Culture. New York, Routledge, 1998 19. Williams E: Beyond Belief. New York, Random House, 1967 20. Johnson PH: On Iniquity: Some Personal Reflections Arising Out of the Moors Murder Trial. London, England, Macmillan, 1967 21. Myers WC: Juvenile Sexual Homicide. San Diego, CA, Academic Press, 2002 22. Kennedy HG, Grubin OH: Patterns of denial in sex offenders. Psychol Med 22:191–196, 1992 23. Shengold L: Soul Murder. New Haven, CT, Yale University Press, 1989 24. Williams AH: The psychopathology and treatment of sexual murderers, in The Pathology and Treatment of Sexual Deviation. Edited by Rosen I. New York, Oxford University Press, 1964, pp 351–377 25. Wertham F: The Show of Violence. New York, Doubleday, 1949 26. Douglas J, Olshker M: The Anatomy of Motive. New York, Scribner, 1999 27. Fogel GI: Perversions and Near-Perversions in Clinical Practice. New Haven, CT, Yale University Press, 1991 28. Court TV’s Crime Libraries. Bind torture kill strangler. Available at: http:// www.crimelibrary.com/serial_killers/unsolved/btk/btk_jump_page.html. Accessed August 22, 2005. 29. Schlesinger LB: Sexual Murder: Catathymic and Compulsive Homicides. Boca Raton, FL, CRC Press, 2004 30. Meloy JR: Violent Attachments. Northvale, NJ, Jason Aronson, 1992 31. Kennedy F, Hoffman H, Haines WA: A study of William Heirens. Am J Psychiatry 164:113–121, 1947 32. Schechter H: Deviant. New York, Pocket Books, 1990

Sexual Homicide

361

33. Dietz PE, Matthews DB, Van Duyne C, et al: Threatening and otherwise inappropriate letters to Hollywood celebrities. J Forensic Sci 36:185–209, 1991 34. Dietz PE, Matthews DB, Martel DA, et al: Threatening and otherwise inappropriate letters to members of the U.S. Congress. J Forensic Sci 36:1445– 1468, 1991 35. Kernberg OF: Sadomasochism, sexual excitement, and perversion. J Am Psychoanal Assoc 39:333–362, 1991 36. Minnesota ex rel Pearson v Probate Court, 309 US 270 (1940) 37. Minn Statutes, Section 26.10 (1941) 38. Gleb G: Washington’s sexually violent predator law: the need to bar unreliable psychiatric predictions of dangerousness from civil commitment proceedings. UCLA Law Rev 39:213–249, 1991 39. Group for the Advancement of Psychiatry: Psychiatry and Sex Psychopath Legislation: The 30s to the 80s. New York, Group for the Advancement of Psychiatry, 1977 40. A Task Force Report of the American Psychiatric Association: Dangerous Sex Offenders. Washington, DC, American Psychiatric Association, 1999 41. Zonana H, Roth JA, Coric V: Forensic assessment of sex offenders, in The American Psychiatric Publishing Textbook of Forensic Psychiatry. Edited by Simon RI, Gold H. Washington, DC, American Psychiatric Publishing, 2004, pp 349–375 42. Kansas v Crane, 534 US 407 (2002) 43. Sreenivasan S, Kirkish P, Garrick T, et al: Actuarial methods for violence and sex-offender risk assessments, in Principles and Practice of Forensic Psychiatry, 2nd Edition. Edited by Rosner R. London, England, Arnold, 2003, pp 752–755 44. Hanson RK: Who is dangerous and when are they safe? Risk assessment with sexual offenders, in Protecting Society From Sexually Dangerous Offenders. Edited by Winick BJ, La Fond JQ. Washington, DC, American Psychological Association, 2003, pp 63–74 45. Corton FA, Marshall WL: Sex as a coping strategy and its relationship to juvenile sexual history and intimacy in sexual offenders. Sex Abuse 13:27–43, 2001 46. Kirwin J: One arrow in the quiver: using civil commitment as one component of a state’s response to sexual violence. William Mitchell Law Rev 29:1135–1219, 2003 47. Tonry M: Crime and punishment in America, in The Handbook of Crime and Punishment. Edited by Tonry M. New York, Oxford University Press, 1998, pp 3–27 48. Janns E: Sex, Violence, and Politics. Ithaca, NY, Cornell University Press (in press)


CHAPTER

11 LEGAL VERSUS CLINICAL VIEWS ON HOMICIDE Diagnosis and Voluntariness

WHEN CONTRASTING THE VIEWPOINTS taken on homicide by psychiatry and the legal system, it must be realized that such a division is arbitrary to some extent and that diversities exist within each approach. For example, in both fields there is a dichotomy between those who take an academic perspective and those who are practitioners. The academic psychiatric critique is that legal jurisprudence has focused on arguing about philosophical questions that are at best of interest only to a small coterie of people. The questions are viewed as not having been resolved over the centuries and as, in fact, possibly not even being resolvable. In response, legal scholars may critique psychiatry as being devoid of understanding the necessity for philosophical and legal distinctions, an understanding they feel is necessary to grasp how a legal system functions in a society. These underpinnings pertain to opinions that in an unacknowledged way, deal with matters such as free choice, lack of capacity, incommensurability, and the inability to understand the nature of an act, to name a few. These omissions are believed inherent in many psychiatric approaches, although the psychiatrists are often unaware of their absence. On the practitioner side, the psychiatrist often finds that in dealing with the perpetrator of a homicide, the attorney is often only seeking conclusory opinions and that psychiatric opinions simply allow the plea bar363

364


gaining process to begin or pressure to be brought on the other side for a resolution of the case. In turn, the practicing attorney finds the psychiatrist all too often just focuses on which psychiatric diagnosis to use and is minimally concerned about understanding what the legal and societal dilemmas are for an individual who has committed a homicide and is charged with some degree of murder. The ultimate naïveté is seen in the position that a psychiatric opinion should be material in disposing of a case. Whether the clinician has emphasized diagnostic classification or psychodynamic formulations, the contrast with the legal system is evident in both approaches. In this section, I exaggerate the contrast between the two viewpoints for heuristic purposes to illustrate the differences. In actual practice, the approaches may not be so diverse.

ISSUE 1: SOCIETAL PROTECTION Legal Perspective The first and most striking difference is the emphasis the legal system places on the primary need to protect society and its citizens from threatening or dangerous people. Hence, the possibility of subsequent or repeated violence is seen as looming in all homicide cases unless a capital punishment is enacted, at which point it is believed total safety has been reached with regard to that particular individual. The theme of protection is preeminent over all other considerations. The presence of a mental disorder that may have possibly been operative in the crime and the type of psychopathology that may have contributed to the homicidal act are secondary considerations. The legal emphasis, then, does not emphasize the well-being of an individual or defendant. It is not even asked whether the perpetrator acted on the basis of some type of derangement. Rather, the focus is a more utilitarian one of striving to maximize the safety of the larger group in a society against those who may or do commit acts of a homicidal nature. Short of this utilitarian component, the legal emphasis might simply be seen as a perspective conforming to a theory of retributive justice, one of the traditional goals of criminal law. Only in more recent times has deterrence been considered as a goal. Rehabilitation was a late goal whose possible ascendancy was very short-lived; in fact, only sporadic, short-lived, and tenuous efforts at rehabilitation as a primary goal of the criminal justice system existed in the course of the twentieth century and at present seem to be a secondary concern. The fragility of rehabilitation as a way of dealing with violent individuals is witnessed in its current downgrading and virtual eclipse. Correspondingly, the reemer-

Legal Versus Clinical Views on Homicide

365

gence of just deserts as the guiding principle of jurisprudence in criminal law has changed how psychiatry participates in the process of dealing with homicidal people. This development is seen most clearly in sentencing practices, which either reflect a shift to a determinate sentencing structure or simply view sentences as a version of “serving time” in which the perpetrator is incapacitated from committing further crimes. An accompanying assumption is that society is best served by the latter approach, with the presumption being that society’s safety is better guaranteed by more frequent and longer incarcerations, if not executions. However, such an argument may be superfluous. It could simply be argued that just deserts demand such a solution. Often it is thought that there is no need to assess what happens to prisoners when they are released after many years of confinement for purely penal reasons. A long incarceration, or an execution, is morally justified as simply being the deserved punishment for an act of homicidal proportions. However, even in the emerging era of prolonged sentences for violent acts, the great majority of individuals are released after a set time period relative to their total sentence. Two issues are bypassed with equal frequency in assessing the effect of such confinement on the released perpetrators: their psychological and motivational status at the time of the homicide and their status at the time of their release. These issues beg the question of whether the sources of violent propensities have been resolved at either end of the process. More profound is the question of whether one should care whether these propensities have been resolved, given that retributory ethics sees these individuals’ imprisonment for the sake of removal from society as a just consequence of their actions.

Clinical Perspective It may seem inordinately idealistic to say that a clinical perspective on the ultimate goals of dealing with homicidal individuals and community safety is needed. Yet it could be argued, in the absence of confirmatory empirical data for either approach, that some type of preventive intervention to deal with the conflicts existing in individuals, their families, or the social milieu that contributed to homicidal behavior might be a more optimal way to maximize community safety in the long run. The focus of such intervention would be to deal with the types of conflict that are present on a personal and environmental level. A more radical view would add that without some fundamental changes occurring in the individuals and a social system that is seen as contributing to the vulnerability of individuals and groups toward violence, continued violence can be expected and pre-

366


dicted. Hence efforts to increase societal protection only by way of increased lengths of incarceration may be doomed to failure.

ISSUE 2: MEASURING SUCCESS OF ATTEMPTS TO DEAL WITH HOMICIDAL INDIVIDUALS A second major contrast between clinical and legal approaches to homicidal behavior is how each approach measures success in dealing with homicidal people and acts of homicide in a society.

Legal Perspective Legal success is determined by low future rates of recidivism for such violence, irrespective of any other goals that might be pursued. However, because homicidal acts are usually connected with subsequent periods of long-term incarceration, it is difficult to evaluate what low recidivism for homicide actually means, given that a prisoner who once committed a homicide is released at an older age after being incarcerated for many years. If we compare the rate of recidivism for homicide convictions with other serious assaultive crimes, the recidivism rate for homicide would look good—that is, how many opportunities for again committing a homicide would there be when someone is serving a long prison sentence? Because a typical sentence may be 30–40 years of imprisonment, the recidivism rate would be quite low. Thus, studies that compare a low rate of subsequent reconvictions for homicides with other major assaultive crimes have a built-in bias toward legal success and possess little validity. Those who have worked in the criminal justice system know that such data are relatively meaningless. Because adequate base rates for determining whether potential behavior has in fact been altered at all are lacking, it is likely that simply some chance variable, such as the passage of time, is the only variable in calculating recidivism rates. The most meaningful approach would be an empirical one in which released murderers are assessed at different follow-up periods and the results are compared. However, such empirical issues may be of little concern to those who believe such an approach may be interesting but unnecessary, given that their priority is community safety. Defining success by reconviction rates also ignores diverse variables in a system, such as the different rates of apprehension and conviction or the impact of plea bargaining. Ultimately, success simply comes to mean incapacitating offenders for long periods of time in some custodial setting. Again, the underlying idea is that such an approach is either a just


367

desert for these individuals’ behavior or that more homicidal types of acts cannot be committed by these individuals within the community (except, perhaps, the prison community) while they are incarcerated. The neatness of such thinking, which amounts to incarcerating people for much of their adult lives, ignores other criteria such as equity, notions of justice, the impact of incarceration on a prisoner’s surviving family members (which often include children), and the attitudes of the general public. Such matters are considered interesting but somewhat irrelevant to the operational definition of success used in a retributive legal approach. For homicidal types of acts that do not result in a homicide but rather in some lesser offense, the fact that most of these people will be released after varying times is also ignored.

Clinical Perspective In contrast to an approach stressing the supposed reduction of recidivism rates, a clinical viewpoint would search to define success beyond confinement. As a beginning, the conditions that produced homicidal behavior would be considered. From a psychodynamic standpoint, an attempt would be made to resolve, or at least ameliorate, the conflicts that have existed either within an individual or between that individual and key people in his or her life. The underlying premise would be that such conflicts are related to whatever elicited such a murderous level of aggression in the first place. Thus a clinical approach that emphasizes interpersonal conflicts would require that these conflicts, whether within a group or in the individual’s social environment, be detected and at least confronted. Success would be measured as the degree to which appropriate interventions lowered the individual’s potential to continue behaving in the same way or, conversely, encouraged the individual to develop alternative ways of handling aggression. A variety of theoretical stances could be taken for assessment purposes. From a behavioral approach, the goal would be to decrease violent behaviors. From a psychodynamic approach, emphasis would be added on resolving the conflicts predating the homicidal act and the subsequent impact such resolution would have on the individual’s behavior. From a more idealized standpoint, the criteria for measuring success in either of these approaches would be whether attempts to actualize the potential of a person or help the individual attain some degree of emotional growth had been even partially successful. The difficulty in making such an assessment would be that for it to be accurate, homicide rates for individuals who did not receive any intervention would have to be measured against those of a control group.

368


ISSUE 3: THE ACT VERSUS THE ACTOR Legal Perspective A traditional distinction between the legal and clinical approaches to homicide has been the focus of the legal system on the act rather than on the actor. Such an approach has some merit. Proponents of this legal approach are interested in the perpetrator’s motives for a homicide only to the degree that intent can be proven. Intent is either used in the sense of common law case development or defined through the process of the legislative grading of offenses. A prosecutor may argue that an act of killing was intentional because it was successfully carried out despite the act being based on a delusion. Many of these legislative definitions of excuse, justification, and provocation have been based on the Model Penal Code of the American Law Institute.1 The danger in this legal approach is that it focuses excessively on the act by statutory analysis, and the nature of criminal propensities within the individual is virtually ignored. From a legal standpoint, the actor’s mental state is subsumed under the concept of mens rea (“guilty mind”), which is used to distinguish degrees of culpability by degrees of pathological indifference to the societal norms. However, assessing mental culpability by the amorphous concept of recklessness on a continuum ranging from purposeful behavior to negligence not only ignores more complex intrapsychic conflicts but also ignores how acts are embedded in social situations. The following are some examples of the types of difficulties that arise when there is an excessive emphasis on the act:2 • Classifying as second-degree murder a death that results from a person driving while intoxicated • Assessing liability for murder based on corporate mistreatment of employees where intent must be assessed in one or more individuals or for the “corporate entity” • Minimizing the mental state required for homicide that can justify the death penalty Other than when raised in the small number of insanity defense cases, motives and stresses on individual personality are customarily considered only at the point of sentencing after conviction or perhaps subsequent to imprisonment. Any interest in the actor as such is even less prominent under the determinate sentencing schemas that are now prevalent. Hence the only time the perpetrator’s motives are considered within the criminal justice system is when a defense of diminished responsibility allied to mental illness is allowed or an insanity defense is raised.


369

Clinical Perspective In comparison, a clinical approach primarily examines the individual actor for a particular predisposing personality type of dysfunction or diagnosis. In this context, motives are seen as not simply a matter of curiosity but as reflecting the personality traits and cognitive and volitional aspects within an individual that need to be understood when a homicide occurs. When a broader inquiry into the mental state of a perpetrator is permitted, such as in a diminished-responsibility jurisdiction, more tolerance is allowed for an inquiry into the actor as well as the act. It is an oversimplification to assume that the legal system is solely interested in the act. To determine mens rea requires an inquiry, at least in part, into the mind of the actor and a search for an intent or a motive (these latter factors are not necessarily equivalent). This line of inquiry is documented by the many legal references to such terms as the premeditation, willfulness, knowledge, and recklessness of someone as well as his or her intentionality.

ISSUE 4: NORMATIVE (LEGAL) VERSUS NONNORMATIVE (PSYCHIATRIC) PERSPECTIVES A fourth contrast between the legal and clinical approaches has an academic overtone. Whereas criminal law is described as a normative enterprise, psychiatry is said to be amoral. In an oversimplified and exaggerated way, criminal law is seen as based on rule violations, whereas psychiatry’s attitude toward rules is seen as laissez-faire. However, in practice, this seemingly neat theoretical distinction is largely invalid because normative, as well as moral, aspects are present in both clinical and legal approaches. For example, substantive criminal law obviously contains norms to evaluate transgressing behavior. Criteria had to evolve if a legal standard was to be created. However, a psychiatric approach is not necessarily amoral. It has its own set of values as to what the norms of behavior should be. The difference is that a psychiatric approach does not confine itself to a “nothing but” assessment of violent behavior by the norms of preexisting criminal statutes. To do so would turn the psychiatrist into a quasi-prosecutor if not an outright moralist. As an example, an acquisitive or quite aggressive act may or may not reflect intrapsychic conflict. It could simply reflect an attempted solution to a problem—adaptive or maladaptive—in conjunction with certain personality traits or developmental arrests. Contrariwise, a psychiatric assessment could also see an acquisitive act as

370


violative of community norms, one that requires the intervention of law enforcement agencies and the judicial system. There is always the hope that the more enlightened the criminal law system becomes, the more interested it will be in how the human personality functions or malfunctions rather than in merely categorizing by graded categories. A less profound, but more prevalent, problem than homicide—that of lying or deception—illustrates the normative versus nonnormative arguments between the two approaches. A strictly ethical position might argue that actions violating the moral integrity of a person can never be justified. Presumably, this stance would mean that a difficult standard exists of the absolute prohibition of lying or deception. The position hinges in part on the way prohibited acts are defined. For example, moral prohibition against killing per se does not extend to all types of killing. Selfdefense is an example of killing that is accepted in almost all societies, as well as killings in war, for which, in fact, the killers may receive medals for their efforts. In an analogous manner, the prohibition against lying does not rule out all deception. Just as self-defense can involve aggression to the maximum point of killing when used to protect oneself against an aggressor, so it has been argued that an act of evasion or deception can justifiably be used to baffle a murderer or rapist to protect oneself or to aid in that person’s apprehension. Why lying or deception is unacceptable might be given different rationales from the legal and clinical perspectives. Customarily, both perspectives would find the behavior undesirable. For example, the legal system might view lying as unacceptable because a system of justice is dependent on people telling the truth. Without such a foundation, or at least the perpetration of its mythology, the system is seen as totally unworkable. Psychiatry also views lying as unacceptable, but for different reasons. The clinical perspective raises questions about what the consequences of deception are in terms of the integrity of the self as well as the effects of deception on human relationships. Not only are human relationships compromised by lying, but there is a distortion in the person’s sense of reality. Lying and deception have developmental components. In this context, children are studied to see how they begin to distinguish between truths and untruths. Part of the distinction lies in a cognitive ability to appraise reality and obey edicts to oneself not to lie. In more dynamic terms, this process would be referred to as internalized prohibitions becoming operative. Descriptive approaches have diagnostic significance when people engage in compulsive lying. For example, the symptom pseudologia fantastica is most often observed in those with hysteroid traits or borderline


371

personality disorder diagnoses, and the question is how aware these personalities are of their distortions. In time, some of these people begin to lose their ability to distinguish truth from falsehood to varying degrees. Cases arise when someone has killed, or attempted to kill, another and subsequently totally denies the act. Because the act was witnessed by several people, at first blush one wonders whether the person is either psychotic or an old-fashioned “cold-steel psychopath.” Yet on clinical evaluation it turns out he or she is neither, such as in the following case example.

Case Example A woman stabbed her estranged husband several times in the parking lot outside his office. Afterward, he managed to get to a nearby hospital and, following surgery, survived. His wife later appeared in the emergency department of the hospital saying her children told her their father was there. She flatly disclaimed any knowledge of the acts and alluded to the possibility that he had done it as a move to help him get custody of their children and make it appear as though she had stabbed him.

Some individuals lie but deny that they are lying, because they see their behavior as something that was not wrong. Alternatively, those with serious religious or political convictions convince themselves that their lying was for a just cause, such as religious truth or the security of the country. More profound clinical questions arise when people with paranoid delusional thought systems are not able to objectively assess their beliefs that they are being persecuted by others and believe that they must act to protect themselves. On that basis, such distortions in truth may lead to violent actions. Is a person lying if he or she kills a loved one and firmly defends the act in the belief that it was a noble or altruistic act? What if similar thinking operates but in furtherance of sincerely held political convictions?

ISSUE 5: EXPLICIT (LEGAL) VERSUS VAGUE (PSYCHIATRIC) APPROACHES A fifth contrast between legal and clinical approaches is that criminal law relevant to homicide is a seemingly well-defined and explicit viewpoint, whereas clinical approaches to dealing with violent people are seemingly vague. The supposed definitiveness in legal approaches is exemplified in the criminal codes that define crimes in terms of the degree of a homicide based on factors such as premeditation, intent, negligence, reckless behavior, and the other factors discussed earlier. These amorphous distinctions are presumed to permeate into the public arena

372


to allow citizens to make rational decisions about the consequences of their behavior. It is supposed that people can then know where they stand if they are contemplating a homicide or similarly violent act. This position is similar to the rational choice model advocated by some criminologists.3 In a vastly abbreviated form, the rational choice model holds that people choose to commit crimes to satisfy certain commonplace needs rather than because they are driven by compelling internalized forces. The choice to act is viewed as requiring a different analysis for different crimes; consequently, crime specificity is a primary factor in the application of this theory. The costs, benefits, and opportunities vary with the different offenses. For a homicide, a series of distinctions would be made ranging from the initial steps toward a homicide to the culminating act. The actor’s final choice would not necessarily be predicated on his or her having adequate information but on acting upon whatever information he or she had, thereby implying a degree of limited rationality. Such presumed rationality often contradicts reality. Legal terms and concepts may be at least as vague as any in the clinical realm, especially in their application to individual cases. Judicial opinions are not likely to provide clarity in delineating their nature. In fact, if all the legal codes and judicial rulings were crystal clear, there would be far fewer contested cases and little need for the endless stream of appellate rulings. Anything that deals with the vagaries of human behavior, be it in a clinical or legal context, necessarily has much attendant vagueness, even within the strict realm of legal concepts. For example, consider the vagueness inherent in discussions about the law regarding attempts at a homicide or the conceptual quagmire of legal and philosophical confusion about a conspiracy to commit a murder. Even though many difficulties are inherent in determining a murderer’s intent at the time of action, this determination remains a key element in bringing legal charges against murderers and deciding what type of plea bargaining might develop.

ISSUE 6: FORMAL FUNCTIONS OF LEGAL VERSUS PSYCHIATRIC SYSTEM In this next section, I contrast the formal functions of a legal system with the functions implicit in psychiatric approaches to a person who is homicidal or who has committed a homicide. The formal functions discussed are as follows: 1) definitions of social relationships, 2) acceptable levels of aggression, 3) mechanisms for dealing with violent people, and 4) redefining relationships in accordance with changing conditions.


373

Definitions of Social Relationships A primary formal function required in a legal system is a definition of relationships among the members of that society, whether a given individual, social group, or political entity is the focus. Within this framework, one function of a legal system is to define what activities will be permitted or not. The goal is to maintain an integration among the activities of people within that society. Clinical approaches also require definitions. However, the definitions of the relationships among people within a certain group (e.g., a family, a broader social unit) are for different purposes. One primary purpose may be to control exploitative activities that can occur within the units. This model can then be extended to larger groups within a society, such as in social interactions, or to relationships within a group for someone who has deviated from the norms. Difficult cases become one of degree. Legal systems need to deal with processes of ostracizing a deviator by way of the criminal process. In contrast, within the clinical approach, a type of social ostracism occurs in which the person is labeled mentally disordered as a preferred explanation for the behavior, which also has an alternative mode of disposition.

Acceptable Levels of Aggression A second formal function in a legal system is setting limits on how much open aggression a given society will tolerate at a given time. In this sense, the legal system is always directed toward the maintenance of law and order. However, this approach does not provide a way of allocating authority or of determining who can exercise physical coercion on others within a group. Some groups, such as police or judges, are given the right or privilege to resort to physical coercion within certain set rules. Sanctions can be used in different ways to achieve particular social ends by coercive measures. The conclusion is not that the field of psychiatry lacks an interest in how individuals display their naked aggression, be it in the violent form of committing a homicide or by one of the subtler means of power used to coerce and manipulate others in interpersonal contexts. Rather, psychiatry’s interest is in the diverse manifestations of aggression and the ego defenses present. Although the rules of interpersonal relationships do not have the same cogency and force of authority as those operating in a legal system, certain people frequently play out the game of controlling other people or groups. Coercive manipulation of any kind always carries the potential to go awry and lead to personal violence.

374


Mechanisms for Dealing With Violent People A third formal function in a legal system is confronting and dealing with troublesome people in a society. One of the troubled groups includes those who commit a homicide. Certainly clinicians are interested in these troublesome people as well, but their interest primarily centers on obtaining adequate and valid diagnoses and on providing treatment. Regarding treatment goals, psychiatry tries to address the conflicts that give these people trouble. Psychiatrists focus on the capacities of these troubled people to live, work, and gain some gratifications in their lives and on how these capacities have become impaired. Clinicians emphasize that their route of dealing with troublesome people is via treatment rather than by what appear to be chance factors. In contrast, vague legal systems of assessing reckless or wanton behavior—either for the purpose of bringing legal charges or for fitting a person into a sentencing grid—seem, at best, like attempts to be pseudoscientific and, at worst, like the manifestation of a politicized agenda. Explanations are offered by psychiatrists for an individual’s behavior to determine the whys and wherefores and how the effect of such an accumulation of unresolved conflict may lead to violent end-points.

Redefining Relationships in Accordance With Changing Conditions The legal system formally attempts to redefine and clarify relationships between individuals and groups under given social conditions that may be in a state of flux. Yet psychiatry may be attempting to do the same thing. In that sense, both systems are required to adapt to change, and neither may be too successful at it. Such changes occur not only in family relationships but as an inherent part of development, in which individuals must come to terms with the broader societal framework that impinges on them. An individual is expected to cope with many changes that may be biological and occur at critical developmental points. Other periods of flux interact with social and psychological variables at crisis points (e.g., graduating from college, entering marriage, dissolving a marriage, menopause, losing a loved one to death, aging). These social variables can lessen the defenses to a major mental disorder and, in that manner, heighten an individual’s vulnerability to violence.4 There is also the need to deal with the ever-changing situations in people’s lives, such as shifting jobs, economic changes, war, and the resurrection of old conflicts that periodically recur and continue to give trouble during the course of a life span. All of these stressful possibilities may require redefining relationships among people, and they can heighten stress and conflict of all types.


375

Interestingly, whereas the legal system derives its working principles by way of postulates developed from the past, as influenced by a particular culture, psychiatry has its own way of being preoccupied with the past. Psychiatry’s concern with people is in terms of the ongoing influences of past events on them throughout their lives. Ultimately, legal, as well as psychiatric, approaches need to focus on the frustrations of individuals and their manners of response. How their frustrations affect relationships with others is often the key to homicides. It may be desirable, albeit too idealistic, to think that relationships can be defined primarily in terms of rights and duties, privileges and responsibilities, and rights and powers of people. The history of human behavior is replete with innumerable examples that indicate the difficulties that arise if we simply limit ourselves to such attempted categorizations. When needed, a public resolution of the types of conflicts that emerge is one job of the legal system. Psychiatric expertise may be called into operation at points where something has gone awry in complex types of relationships. Ideally, this request is made before problems have become public, but the call may come afterward, when the conflicts have been enacted in the public arena.

USE OF PSYCHIATRIC DIAGNOSES IN THE LEGAL SYSTEM In addition to the judicial system and the other branches of government, many other disciplines have an interest in the subject of homicide. Psychiatry has a long history of emphasizing that a valid diagnosis is important. Yet some would question whether it matters if a diagnosis is made in homicide cases. Many articles and books have been written about diverse types of killings without any references to psychiatric assessment or diagnoses of the perpetrators whatsoever. Similarly, most homicide trials may take place without any reference to the psychiatric status of the perpetrator. A discussion of the difficulties and merits of psychiatric diagnoses in the legal system follows.

Diagnostic Labeling Versus In-Depth Psychiatric Assessment Some criticize limiting psychiatric contributions to merely providing a DSM-IV-TR5 diagnosis. Even within the confines of clinical work on homicide, it is surprising how much emphasis is placed simply on addressing the issue of whether the accused is psychotic or not, without appreciating the multiplicity of factors that overlap in diverse diagnoses. Discussions may wander off into specific legal or social contexts

376


of the act, thereby ignoring psychiatry’s understanding of causation, experience, classification, and therapy.6 Unfortunately, such omissions are present in most reports to courts and attorneys in the assessments of homicidal defendants, and these omissions are similarly witnessed in psychiatric testimony at trials. The need for careful thinking when applying psychiatric diagnoses to legal questions about homicides is clear. Equally clear is that an evaluation of individuals should never be viewed as complete simply when a descriptive diagnosis is rendered. There is also the principle that diagnoses are not diseases, which needs to be kept in mind in these situations.7 In any type of psychiatric assessment, it is implicit that a key variable is the capacity of the person to act, or to be able to refrain from acting, in certain ways. To exercise choice about one’s behavior reflects a capacity to weigh alternative choices. Trying to reconstruct the mind of the murderer means that whoever is doing the reconstructing is almost always looking backward in time. (One exception would be if a researcher were attempting to assess a high-risk group and make predictions as to who might be subject to homicidal violence sometime in the future, or make predictions for some type of preventive assessment.) When legal questions are raised about an individual’s mental state during the commission of a homicide, in the absence of guidelines, psychiatrists may take the liberty of expanding on their own ideas regarding the person’s mental state. However, this historical state may or may not correlate with the person’s existing or nonexisting incapacities. A psychiatrist, analogous to other medical specialists, addresses the question of what limitations exist in the capacities of a given person and why they exist. In a broad sense, in every appraisal of an act of homicide, a moral question always lurks: whether, and to what degree, a person should and can be assessed as blameworthy. Of course, there are also clinical questions: Has a medical condition impinged on the capacity of a person? Has some adverse type of social situation been present? What has contributed to irrational cognitive processes? Have cumulative traumatic episodes been operative in the person’s life in the prehomicidal period? A common misconception among nonpsychiatric professionals, including some attorneys, is that the psychiatrist performs in a manner analogous to other medical specialists, in which the specialist (e.g., a surgeon, an internist) performs a procedure on a patient or assesses the patient’s condition through inspection, palpation, or tests, after which the specialist arrives at a diagnosis. Apart from questions about the degree of validity of applying such thinking to other medical specialties, this approach bypasses the essential reasoning processes followed in psychiatrically assessing a patient. These processes include reviewing information from several


377

sources, such as reports from places of employment, schools, and hospitals; medical histories; probation reports; statements of witnesses; results of psychological testing; investigatory reports by police officers and detectives; and so on. Only when the patient’s reported signs and symptoms are put together with his or her social and historical background can the diagnostic impression, and possible incapacities contributing toward homicidal violence, be adequately assessed.

Limitations of a Diagnostic System in Legal Settings In the situation of a person charged with a homicide, specific legal definitions operate apart from questions about the presence of syndromes, disorders, or diseases. Many times statutes or legal cases refer to the presence or absence of a mental disorder or disease without defining the terms. An insanity statute may require the presence of a mental disease and a “defect of reason” that may or may not be present in a particular mental disease. In addition, different jurisdictions have interpreted these medicolegal terms differently. Implicit in some definitions of mental disease may be the idea of limitations on the ability of a given individual to form an intent to carry out a certain act. The psychiatrist would then need to elaborate on what these limitations are. A common mistake encountered in the legal arena is to assume that every disorder listed in DSM-IV-TR is a disease. Even if a psychiatrist wants to take a position that a particular disorder is a disease, that disorder may not necessarily indicate an impairment in the capacity of a person to make choices. In and of itself, a diagnosis can never establish responsibility, or its lack, for a homicidal act. Confusion may be multiplied at the junction between clinical diagnoses and legal questions. First, the purpose of the diagnostic manuals, such as DSM-IV-TR, is not to offer explanations for behavior. Only a few of the diagnostic categories are related to etiology, such as delirium, dementia, and amnestic and other cognitive disorders, which are presumed to be related to a medical condition and/or a substance. Even with these few categories that imply an etiology, a particular behavior, such as homicide, is not explained without many additional intervening factors. To say that Mrs. Smith, who carries a diagnosis of a senile dementia, shot and killed her husband because she had dementia is an insufficient explanation at best. At worst, it may also be a flatly erroneous opinion offered as a possible explanation. For a particular diagnosis to be used as the explanation of a homicide, a step-by-step connection between the signs and symptoms of a particular diagnosis up to the final act of a killing must be made. An explanation of homicide also means that various aspects of personality

378


functioning must be assessed in terms of their functional capacities. These capacities might be impaired in some areas of ego functioning but not in others. The dysfunctional areas, in turn, need assessment to determine whether, in fact, they are related to homicidal behavior. In making such an assessment, all sources of information need to be used. Even in legal cases not involving crimes as serious as homicide, examples of misattributions abound. As an example, a problem may arise with a series of lewd phone calls made in a neighborhood. The assumption may be that a rehabilitated former patient with a diagnosis of schizophrenic disorder is the culprit because when he had once been in a psychotic state, he made threatening, although not lewd, calls to his family. If a homicide occurs, this type of slippery reasoning emerges in the search for a suspect; in fact, such thinking is pervasive in police investigative work. Another example is when a sexual homicide occurs in a neighborhood; in collecting suspects, the police gather the names of not only anyone who has ever committed a sex offense but also those who have had a history of mental illness. Diagnostic manuals have tried to make a disclaimer for their use in reaching legal conclusions by noting that the classification of mental disorders is simply a reflection of a consensus of current formulations in an evolving field of knowledge. They have referred to the clinical and scientific considerations involved in categorizing conditions as mental disorders as not being wholly relevant to legal judgments. Issues of responsibility, disability determination, and competency are not resolvable simply by having a diagnosis. On the other hand, if courts are to ask psychiatrists for their expert opinions about the state of mind of individuals who kill other people, it is unavoidable that some type of diagnostic nomenclature be introduced. Diagnosis provides a classification of mental disorders with criteria that certain people may meet. It is not an overall classification of people, nor does it offer precise boundaries for what is or what is not mentally disordered behavior. Such limitations further complicate the use of psychiatric nomenclature in legal work. The most a classification system can do is conceptualize a mental disorder as a clinically significant behavioral or psychological syndrome or pattern that is associated with pain, distress, disability, death, or loss of freedom or that raises the possibility of such phenomena. Deviant behavior of a political, religious, or sexual nature or conflicts that are primarily between individuals and their society or other societies are not conceptualized as mental disorders. The psychiatric goal is to conceptualize the behavior as reflecting a behavioral, psychological, or biological type of dysfunction. Although the search to understand political killings may benefit from some psy-


379

chiatric input, such input is insufficient to explain completely the killings connected to political paranoia or terrorism.8 Although a diagnostic manual may facilitate a classification to aid in the accumulation of scientific knowledge about a particular category, and perhaps the best related treatment, it does not address per se the question of capacities that have been mentioned as crucial to the operation of the legal system. However, the converse does not hold, in that if one starts with a basis for some incapacity, this basis presumptively raises questions about the possibility of a mental disorder contributing to such a state, but no more. Certain complaints about limitations on thinking may signify to a psychiatrist a differential diagnosis. For example, a retrospective history taken after a homicide, in which there have been antecedent behaviors of irritability, “flying off the handle,” difficulties in relating to others at work or at home, brooding anger, and the expression of needs for vengeance and justice, may suggest that the significance of these behaviors be assessed from a psychiatric perspective. Similarly, specific past expressions of thoughts about carrying out a suicide or homicide indicate the need for assessment. Listings of such complaints can only suggest possible descriptive diagnoses. For example, a limitation of functioning connected with a diagnosis of a bipolar or schizophrenic disorder may be suggested, but the diagnosis itself lacks specificity about what functions may be impaired. Achievement of specificity requires an individual assessment and determination. Many people carry ominous diagnoses of a psychotic disorder, yet they continue to function well in different areas of their lives. Such insights are as valid for schizophrenia as for any other diagnosis in which a false assumption is often made that an ominoussounding diagnosis by itself carries an increased homicidal risk. Whether any such grouping, assuming valid diagnoses, carries such a risk is an empirical question. In reality, this question is only now beginning to receive the scientific scrutiny that has long been needed.9 What raises more complicated questions is the generic diagnostic grouping of personality disorders and their relationship to homicides. The diverse personality diagnoses cover a host of conditions, many of which are associated with unacceptable social behavior varying from impulsive acts, to thefts, to sexually conflicted acting out, to being a factor that contributes to a homicide. Even more troubling is when the diagnosis of posttraumatic stress disorder (PTSD) is offered as an explanation, or perhaps even as an exculpation, for homicidal violence (this issue is discussed at length in Chapter 7, “Masochism and Homicide”). The ultimate expression of the posttraumatic situation is where the perpetrator of a homicide is portrayed as having been inadequate and

380


helpless at some earlier point in his or her life, and the homicidal act is conceptualized as occurring because the person could see no other alternative available at that time except to destroy the person who was perceived as being the oppressor in his or her life. Such reasoning raises probing questions about individual responsibility. Post hoc, ergo propter hoc fallacious reasoning may surface. Whoever is retrospectively seen as having been the earlier oppressor is, after the homicide occurs, interpreted as the heretofore undisclosed cause of the violence. This type of reasoning invites conclusions that the person killed is the one actually responsible for the homicidal violence. The perpetrator would not be seen as responsible because of the preexisting trauma that had been inflicted on him or her. For example, an adolescent whose father beat him over a period of years is then seen as responding to the beatings when he kills his father or someone else at a later point. Another example is when a homicide occurs and the past issue of childhood sexual abuse is raised as the predisposing, if not causative, factor. A related problem is that simply by virtue of applying a diagnostic label, the possibility for a harmful act may be suggested. The question becomes one of how far society and the legal system wish to go in accommodating the violent behavior of a person. Although the person may be idiosyncratically homicidal, it is treacherous to argue for a causal view that the person lacks the capacity to control his or her actions because of a particular diagnosis or because of some antecedent traumatic events. The clinical error discussed earlier is sometimes referred to as the diagnostic fallacy. The fallacy consists of coming up with a diagnosis, whether valid or invalid, and then making a leap from it to an assumed explanation. The most frequent example occurs with some type of psychotic diagnosis, such as schizophrenia or bipolar affective disorder. For example, a presumption may operate that a proposed diagnosis, such as schizophrenia, paranoid type, by itself carries an obvious explanation as to why a killing took place. In fact, the diagnosis may not have any direct relevance to the killing any more than some other diagnosis might. In its crudest form, this chain of reasoning offers a conclusory opinion that someone has a schizophrenic diagnosis and, therefore, a causal relationship to an act of homicide exists. The result begs the question. Instead, what is needed is to examine the multiple factors that may be operating, rather than concluding that “Mr. Jones is schizophrenic, which is why he killed the intruder in his apartment building.” One obvious possibility is that schizophrenic individuals may have burglars in their dwellings and may decide to shoot at them, just like any other citizen. Fallacious diagnostic thinking of this sort seems to arise more of-


381

ten in the context of hospital settings, where a diagnosis carries “excess baggage” in terms of seeming to explain all manner of things about a person, such as why the person does not eat the food offered, objects to wearing a hospital gown, will not participate in group therapy, will not take medications, and so on.

VOLUNTARINESS The question of the voluntariness of behavior in general is an old and complex one for psychiatry and law and has preoccupied philosophers over the centuries.10 It continues to be debated without resolution and recurs repeatedly with regard to some psychopathological conditions. Contemporary psychiatrists and psychologists do not have an answer to the free will problem any more than do philosophers. The most that can be argued is that certain types of mental disorders impinge on the degree of freedom a person has about choosing between options. To conclude that a person has no choice in violent acts, and that he or she had to kill someone, is an undemonstrable position. The underlying question is the degree to which choice operates with respect to actions, in contrast to a viewpoint where the acts occur because of causes beyond the person’s choice. The “compatibility” question of philosophers is how free will can be consistent with a determinist viewpoint.11 The determinist argument is that a feeling of conscious will is actually caused by what occurs in the brain or “the mind.” Mind itself is subject to philosophical debate because mind may be equated with brain to eliminate the idea of a “ghost in the machine.” The dilemma posed is that if neuropsychological mechanisms are responsible for actions, what role is left for the action of a conscious will? The conclusion is that we may feel we are consciously willing something to occur, but it is rather actions that happen to us. Conscious will would thus be an illusion, although it gives us a sense of personal responsibility.12 Pinker,13 a cognitive psychologist, takes up biological and environmental arguments that try to eliminate a determinist position. Even if neuroscientific knowledge is probabilistic, it does increase our ability to predict. Pinker views efforts to negate determinism by referring to the probabilistic factors in brain functioning or evolutionary history as giving little comfort to those who wish to see choice as free. Those who hope to banish the argument of biological determinism can also take little comfort in the alternative of an environmental determinism that is based on the idea that bad things that happened to people are the cause of their later bad acts. This deterministic spin refers to a multitude of causes that supposedly determine acts and then later surface in the

382


form of environmental explanations, such as the abuse excuse, the Twinkie defense (consuming huge quantities of junk foods, which causes some type of allergic reaction equivalent to a seizure), black rage, the impact of pornography, media violence, rock music, and various cultural influences on violence. The ultimate irony would be seen when a person is shot and the explanation offered is that it is undoubtedly simply due to some childhood trauma the shooter experienced. Recent neuroscientific work may well have a bearing on these issues of determinism. As an example, the question persists as to why some boys who are maltreated and abused grow up to develop antisocial behavior whereas other abused boys do not. One line of research has found that a key differential may be that the children most capable of handling the maltreatment are those with the gene that encodes the neurotransmitter-metabolizing enzyme monoamine oxidase A (MAOA), which can moderate the effects of maltreatment.14 The maltreated children who had low activity levels of MAO-A were not able to constrain the effects of their early treatment and were more strongly predisposed to adult violence. The theory is that maltreatment in early years alters neurotransmitter systems; these alterations persist over time and later influence aggressive behavior. These types of genic influences on behavior raise many questions about the degrees of free choice. However, as many have pointed out from diverse disciplines, attempting to arrive at explanations for why an act of violence has occurred is separate from excusing the conduct. In other words, explanations are not equivalent to excuse or exculpation. Scientific explanation seeks causal networks, be they biological or environmental. It operates within the framework of seeking regularities in behavior and perhaps predictability. Without such a framework, behavior would be chaotic and thus would itself be opposed to assigning responsibility. In behaviorist language, contingencies of punishment are believed to deter some behaviors. However, blameworthiness is assigned if acts are committed. This assignment of blame is based on the legal system’s dependence on the idea of individual responsibility. The position is derived from Kant’s idea of autonomous beings that are responsible because they exercise a choice, whatever the social or power context of the behavior may be.15 It is again striking how this type of legal thinking is similar to psychiatric ideas about behavior. When there are difficulties in control, psychiatry can perhaps add something about how autonomy can become compromised in seriously disturbed individuals who commit homicidal acts. However, the legal system operates from the position that individuals are conscious, intentional creatures with volitional capacities. Those who commit a homicide un-


383

der the influence of a mental disorder may be found less blameworthy legally because their rationality is impaired, but not because their mental disorder plays a causal role in explaining the conduct.16 The difficulty is that when viewed from a strictly clinical perspective, all mental disorders, including the even broader area of conflicted behavior, impinge on one’s ability to make choices. The striking example of a person with schizophrenia who has command hallucinations from God to kill someone is one extreme, but even in such cases most of these individuals do not seem to obey the command. Those with schizophrenia of the paranoid type who have delusional beliefs that their thoughts can be read by others can also be seen as having a limitation on their rationality and hence choice of options. Personality disorders offer enormous complexities in terms of assessing voluntariness. Factors such as temperament, character, and traits limit the degrees of freedom. It is interesting that although clinicians often conclude that individuals with a psychotic diagnosis have a compelled quality to their actions, clinicians are equally resistant to such possibilities for character problems, particularly those related to a diagnosis of antisocial personality disorder. Instead, they more often assume that individuals with this personality disorder have total freedom of choice with respect to all rule violations, including the most egregious self-destructive acts. Matters become more complicated when individuals with such a diagnosis who kill do so with some sense that the act was justified, even in a moral sense.17 The act of killing may be seen as a form of moral revenge for someone having violated their rights or honor. Yet in the absence of a delusional system, the acts are interpreted quite differently. When explanations for homicide are offered in terms of a causalitydriven clinical model, the accompanying reasoning may more readily suggest exculpability for the person’s acts. The fallacious leap from explanation to exculpation frequently operates in material presented in courtroom testimony. The converse also holds when a satisfactory causal explanation for a homicide is not available; the assumption may then be made that the act is either a rationally motivated act or that the killing is one without a motive, meaning it lacks an explanation. A person who kills another person at a gambling table when tempers flare is assumed to have acted out of anger; that person’s action, however, is seen as a rational act, just like the myth of the cowboy heroes of the Old West who decided to draw and shoot during a card game.18 However, if the killing seems to have an irrational quality, such as a person walking up to an unknown person on a street corner and plunging a knife in his or her back, the act is presumed to be irrational, and exculpatory modes of thinking quickly come into play.

384


The disciplines of psychiatry and psychology have no basis to partition motives for behavior into a rational or irrational dichotomy to assess responsibility. In fact, even to use such a dichotomy clouds the assessment of a person’s functioning. For example, a young male wandering into his parents’ bedroom and shooting them would seem irrational because most people do not behave that way. Yet if logical reasons such as revenge begin to be adduced for the behavior, the act may begin to seem more rational. Either way, a diagnosis per se cannot be the arbiter of culpability. A caveat is needed when attempts are made to predict homicidal behavior from a particular diagnosis. The danger in such thinking is that it is made to seem as if the act flowed in a deterministic manner. Some researchers have attempted to show that certain diagnoses have a higher association with particular types of behavior than other diagnoses, such as depression with suicide or antisocial personality and bipolar disorders with more risk-taking behavior. Therefore, questions arise about whether these diagnoses carry a greater likelihood of a homicide occurring. Again, analogical thinking is present in hypothesizing that because suicide is self-destructive, then those who commit suicide also carry a greater propensity for other types of destructiveness. Are those who more easily violate norms or more often take risks more likely to kill as well? It could be argued analogously that certain states of despair, anger, and impulsivity give a person a greater disposition to act irrationally. Although such reasoning holds a degree of metaphorical appeal, great caution is needed when applying metaphors to prediction models in individual cases. If valid actuarial data could be obtained, then correlations between diagnoses or certain traits and homicidal behavior could be made with higher probabilities. However, it would be treacherous to infer that any one member of a group exhibiting such a diagnosis or set of traits would in fact behave in a homicidal manner. Yet that is exactly the position courts or legislatures have taken when they have adopted rules that hold clinicians accountable, in retrospect, for making or not making such predictions. Out of such thinking have come cases holding that clinicians have a duty to warn, or protect, others about the possibility of dangerous acts by patients. The Tarasoff case in California produced a progeny of cases and legislation in this area.19 Such treacherous thinking is also seen in the expanding area of malpractice cases against mental health professionals who are seen as having failed to predict a suicide or homicide; in those cases, the allegation is that the patient’s standard of care was violated.20 Although hindsight with respect to certain capacities in a person may be closer to 20/20, the diagnosis itself


385

before a homicidal act can never be a highly reliable predictor of a homicide, nor can the existence of certain traits in a personality. At most these traits can tell us that a certain combination of factors, such as a psychosis with a comorbid substance abuse problem, increases the risk of a homicide occurring. One of the most difficult areas in assessing voluntariness is the burgeoning number of cases attributing a homicide to the past occurrence of some sexual or physical abuse (see discussion of the PTSD model in the previous section, “Limitations of a Diagnostic System in Legal Settings,” and in Chapter 7, “Masochism and Homicide”). The original model focused on antecedent abuse in childhood, leaving the person vulnerable to committing a violent act because of the unresolved effects of PTSD. The model has now been extended to abuse occurring in adulthood as well, or to harassing behavior that may also be claimed to be a causal antecedent to the unavoidable outcome of an outburst of violence. The abuse or harassment is posited as the antecedent event that impairs the capacity of the person to control himself or herself in a subsequent act, such as the murder of the abuser or a substitute for the abuser; hence, the subsequent act is proposed as not voluntary. Note that in DSM-IV-TR events such as simple bereavement, chronic illness, business losses, and marital conflict are not classified as traumas. Instead, these would be classified as factors that lead toward an adjustment disorder diagnosis. The stressor for the traumatic state in DSM-III-R21 was something markedly distressing to almost anyone and experienced with intense terror and helplessness. By the time of DSM-IV,22 the criteria were expanded to a traumatic event in which a person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury, or a threat to the physical integrity of the self or others, and a response thereto involving fear, helplessness, or horror. The symptoms involve reexperiencing the traumatic event, avoiding stimuli associated with the event, or experiencing a numbing of general responsiveness and increased arousal. A subsequent homicide is then explained with the reasoning that the previous trauma was of such a nature as to pose a serious threat to one’s life or physical integrity. In the context of reexperiencing the traumatic event, the person may be reported as having recurrent dreams or intrusive recollections of the past event or can enter dissociative states lasting from a few seconds to days in which the events are relived. Within this context, the person supposedly behaves as if experiencing the past traumatic event at the time of a killing. Among those changes the person experiences is a change in the degree of aggression from increased irritability to fears of losing control.

386


In considering the connection between homicide and predating traumatic events or situations, one must consider that the great majority of people exposed to an event that is markedly distressing may show certain reactions for a brief period of time but no more. Subsequently, extreme degrees of responsiveness are present in only a small number who show the clinical picture of PTSD. When a homicide occurs in the absence of psychosis, a host of possible diagnoses may be offered besides PTSD, such as dissociative identity disorder, dissociative state, depersonalization, sleepwalking violence, fugue, and a variety of what are claimed to be actions occurring in states of altered consciousness. In all of these, the implication is that the freedom to act otherwise has been lost. It could be argued that a spouse who kills her mate because she fears continued abuse from him is prevented to a degree in exercising free will and simply leaving the relationship because she fears her husband will find her later; to a woman in such a situation, killing might seem to be the only rational choice she can make to prevent her experiencing a harmful outcome. When a person is involved in an emotionally painful relationship that is repetitious and not resolved over time, he or she may begin to think of homicide as the only solution. If this person is in therapy, the matter of exerting a choice may be stressed as part of the workingthrough process. Yet once a homicide occurs, it is as though no choice had existed. The interpretation may then be offered that the person reexperienced an old situation in a dissociated state when he or she committed the homicide, as in the following case examples.

Case Examples A male was approached in a washroom by another man who was soliciting homosexual contact. The male claims that he became overpowered by fears from a previous, related situation and, feeling powerless, killed the man to protect himself. He later asserts that he had to kill the man because the scene in the washroom aroused the mental state he once had when he was abused as an adolescent.

Note that in this above case, based on his own explanation, the man had sufficient control to make a choice of avoiding a state of powerlessness. A war veteran exposed to killing in combat got into repeated fights in bars; these fights often, although not always, occurred when he was intoxicated. He claimed that the fights occurred when he was reexperiencing the state he had when engaged in combat. The lingering effects of unresolved PTSD left him vulnerable to misinterpreting current situations. Alcohol would be functioning as an added facilitator here.


387

These predicaments raise troubling questions. First is the validity of whatever diagnosis is used. Next is the difficulty in confirming what is alleged to be the trauma. Although in some cases the trauma can be confirmed through outside evidence, in many cases it never can be. There is often an extreme degree of self-justification involved when diagnoses that have not been diagnosed before a homicide are raised in court cases, which is usually the case.

CONCLUSION: THE RATIONALITY OF THE IRRATIONAL Where do these appraisals leave us with respect to psychiatric diagnosis, choice making, and acts of homicide? They demonstrate that rendering a valid and reliable diagnosis of a person is valuable but has limitations in terms of explanations for homicides. Assuming that the components associated with a diagnosis can be confirmed as operating at the time of the homicidal act, what is still lacking is an explanation of how the mental impairments that were present actually contributed to the act of killing. The key is not only what past factors were operating in a particular person and the person’s disposition toward acting in a certain way, but what triggered the act of killing at a particular time and place. The ultimate question is what psychological processes coalesced at the time of the homicidal act. For some, such a formulation seems too rational. It implies that a person about to commit a homicide reflects, makes choices, and then carries out the act. In fact, this framework would neatly correspond to the legal definition of a premeditated first-degree type of murder. Although rational components are often mixed in with the irrational in homicides, to understand an act beyond a superficial level requires much more than a determination of whether someone could make a choice. Not only are there many perpetrators who cannot sincerely give a rational explanation for their acts, but also many give sincere explanations that are not sufficient. In this setting, if the psychiatrist is to go beyond a layperson’s level of explanation, he or she must look for impairments in personality functioning as well as unresolved conflicts and maladaptive solutions. Depending on how far one wishes to give license to hypotheses that cannot be disconfirmed, interpretations of behaviors are sometimes offered on a metaphysical level. In the framework of ferreting out preexisting conflicts and maladaptive responses after a homicide occurs, the clinical investigator must seek a pattern of behavior that evolved into a homicide as a final outcome. A variety of choices may have been made by the person on such a pathway, with varying degrees of consciousness about the significance of each of these choices being operative at the time. I illustrate this point in the following example (indicated by numbers in brackets):

388


In a pattern not uncommonly observed in dysfunctional marriages, [1]the parties realize that they have serious problems after an extended period of time has passed. [2]They may have sought help from the usual variety of counseling services available and, perhaps, have had some separations. [3]For a particular subset of couples, a sense of their being inextricably bound and not able to separate emerges.23 Yet they cannot continue to live with each other in any type of peaceful coexistence. [4]In this setting, one of them kills the other—perhaps during an argument, while drinking excessively, or in the context of some comments being misinterpreted. [5]The perpetrator is subsequently overcome with grief and remorse and protests that he or she did not mean to kill his or her loved one.

In this situation, an endless series of choices have actually been made by both parties without any conscious awareness that they are on a pathway toward a homicidal collision. It is not that they fully understand their motives and unconscious basis for some acts, but rather that they are unaware of choices being made and the enhanced risk of violence. Another situation frequently arises in the context of a mentally ill person’s laboring with some psychotic cognitive processes. This type of thinking manifests itself in regressive episodes where misinterpretations begin to be made, partial conclusions are drawn, or an isolated event is seized on and given exaggerated significance. Such thinking is not always present, nor is it always fragmented. At times, it is kept in abeyance, perhaps when the person is maintained on appropriate medication. For a variety of reasons, many of these people elect to stop taking their medicine. In time, their cognitive disturbances recur and, on that basis, they begin to act on some of their misinterpretations, even to the point of a homicidal act. In these cases, questions arise about the mental capacity of these individuals, given that they actually decided to stop taking the medication, which resulted in a series of other choices being made that ultimately resulted in a homicide. In very few cases does anyone have more than a scintilla of knowledge about the long-range consequences of choices that he or she is making. Some usually insist on a rational basis for all of their decisions. The more conflicted a person is in terms of psychological disturbances (e.g., having a gross impairment in his or her thinking, being disgruntled, having attentional problems, experiencing intrusive thoughts, making personalistic interpretations, having a lack of empathy with others’ feelings, being preoccupied with needs for revenge, experiencing delusions, being caught in dejected moods with concomitant struggles), the greater is the predisposition to a violent outcome. Events that involve individuals with such striking impairments in their functional capacities are the easier ones to reconstruct retrospec-


389

tively. More difficult are those situations in which individuals exhibit seeming rationality and intactness before a homicide and perhaps afterward as well. It is not that they are attempting to feign being disturbed but rather that they have become convinced about the desirability, if not the necessity, of their behavior. One example is the person who harbors private delusional beliefs about a neighbor who is seen as somehow engaging in behavior harmful toward him or her, and after years of such brooding, the person acts on that delusional belief. The danger arises from processes in these individuals when they block out certain, more correct perceptions or deny that other interpretations are possible as to what the events in their environment mean. What happens in these cases is similar to the following examples. • A subtly paranoid individual denies information that would disconfirm inferences that he or she is making. • A depressed patient becomes convinced that this is the worst of all possible worlds (turning Candide on its head) and therefore concludes that some action needs to be taken, perhaps to spare everyone more suffering. • An obsessional person cannot stop focusing on how a realtor sold him a house, which he has concluded was not only overpriced but also has many defects. He progresses to brooding and depressive symptoms. After repeated efforts to gain relief from the realtor are rejected and his attorney tells him that his court case will come up in a matter of years, he feels that the only way to regain his sense of composure is to confront the realtor. The realtor is asked to sign a document that he will reimburse the buyer for a given amount of money. When he refuses to do so, the buyer shoots him. • A helpless, senile individual gets word that his wife has developed a malignancy. He concludes that he wants to spare her any suffering. As a result, he immediately prepares to take action and shoots his wife. • A manic patient with a bipolar disorder overcelebrates, becomes intoxicated, and proceeds to drive so recklessly that the car spins out of control onto a sidewalk and kills a pedestrian. These examples all raise questions about the ongoing mental functioning of such people at the time of a homicide and how the impairments involve, to different degrees, cognition, perception, the capacity to make choices, ego dominance over impulses, and volitional components. In many of these cases, the person is distraught and suffering to some degree. Some remain convinced that they behaved in a rational manner given their situations. In some of these cases, psychiatrists

390


would diagnose a type of psychotic disturbance, yet other individuals have personality disorder disturbances, and some have comorbid disorders. Although choices are still being made, these difficult cases raise profound questions about the choices not being made freely, or certainly not as freely as someone would make without these handicaps. These examples raise fundamental issues and contrasts between the legal and psychiatric viewpoints when a killing occurs.

REFERENCES 1. American Law Institute: Model Penal Code, Part I: General Provisions. Philadelphia, PA, American Law Institute, 1962 2. Weisberg R: Criminal law, criminology, and the small world of legal scholars. Univ Colo Law Rev 63:521–568, 1992 3. Cornish DB, Clarke RU (eds): The Reasoning Criminal: Rational Choice Perspectives on Offending. New York, Springer, 1986 4. Blugra D, Leff J (eds): Principles of Social Psychiatry. Oxford, UK, Blackwell Scientific, 1993 5. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000 6. Littlewood R: Against pathology: the new psychiatry and its critics. Br J Psychiatry 159:696–702, 1991 7. Mindham RHS, Scadding JG, Cawley RH: Diagnoses are not diseases. Br J Psychiatry 161:686–691, 1992 8. Robins RS, Post JM: Political Paranoia: The Psychopolitics of Hatred. New Haven, CT, Yale University Press, 1997 9. Monahan J, Steadman HJ, Silver E, et al: Rethinking Risk Assessment. New York, Oxford University Press, 2001 10. Honderich T: How Free Are You? The Determinism Problem. New York, Oxford University Press, 1993 11. Honderich T: After compatibilism and incompatibilism, in Freedom and Determinism. Edited by Campbell JK, O’Rourke M, Shier D. Cambridge, MA, MIT Press, 2004, pp 305–321 12. Wegner D: The Illusion of a Conscious Will. Cambridge, MA, MIT Press, 2003 13. Pinker S: The Blank Slate: The Modern Denial of Human Nature. New York, Viking, 2002 14. Caspi A, McClay J, Moffitt TE, et al: Role of genotype in the cycle of violence in maltreated children. Science 297:851–854, 2002 15. Norrie A: Punishment, Responsibility and Justice: A Relational Critique. New York, Oxford University Press, 2000


391

16. Morse SJ: New neuroscience, old problems, in Neuroscience and the Law. Edited by Garland B, Frankel MS. New York, Dana Press, 2004, pp 157–198 17. Katz J: Seductions of Crime. New York, Basic Books, 1989 18. Slotkin R: Gunfighter Nation: The Myth of the Frontier in TwentiethCentury America. New York, HarperPerennial, 1993 19. Tarasoff v Regents of the University of California, 17 Cal 3d 425, 551 P2d 334, 131 Cal Rptr 14 (1976) 20. Beck JC (ed): Confidentiality Versus the Duty to Protect: Foreseeable Harm in the Practice of Psychiatry. Washington, DC, American Psychiatric Press, 1990 21. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987 22. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994 23. Cormier B: Depression and persistent criminality. Can J Psychiatry 11 (suppl):208–220, 1966


CHAPTER

12 12 HOMICIDE IN THE TWENTY-FIRST CENTURY Where Knowledge Is Needed

TO STUDY VIOLENCE only by way of the extreme examples of homicides may be too restrictive a concept for research purposes. In earlier chapters, I noted many acts short of homicide that are equivalent in a psychiatric sense to the legal categories of attempted murder or aggravated assault. They not only involve the same external circumstances but also the same intrapsychic factors. Furthermore, diverse types of conflicts can lead to different types of violence in which one possible outcome could be a homicide. Conversely, a homicide can result from situations in which all the ingredients for a homicide seem lacking except the final end result. These types of acts ordinarily do not result in a killing, yet a homicide results from a chance ingredient. One major problem in understanding homicide is that efforts to study extreme violence have originated in several different disciplines. Although on the surface it would appear that the knowledge gained from all these perspectives would, when combined, give a more comprehensive, overall picture of homicide, in actuality little integration of knowledge has resulted. Historically, homicide has been researched from legal and social perspectives; psychiatry’s perspective is a relative 393

394


newcomer. Researchers from the fields of physiology, biology, and the entire scope of the social sciences have all taken an increasing interest in homicidal violence. Efforts to comprehend and develop research strategies for studying homicidal types of violence often seem to become so far extended that their focus becomes indistinguishable from a study of aggression. This blurring of the distinction between aggression and homicide can create confusion in interpreting results, given the wide variation that can exist between aggression per se and homicide.

DETERMINING THE NATURE AND EXTENT OF HOMICIDE IN CONTEMPORARY SOCIETY As mentioned in Chapter 1 (“Epidemiological Aspects of Homicide”), one way to approach the study of homicide is to use epidemiological methodology. However, criminologists have long been aware of the risks in using official statistics from diverse agencies as a basis for offering valid conclusions about the extent of any criminal behavior. This caveat also holds true for something as seemingly discrete as homicidal behavior. Homicidal behavior is more prevalent than the actual clearance rate of completed homicides, as noted in reports such as the Federal Bureau of Investigation’s Uniform Crime Reports. Some of the problems in obtaining valid statistical assessments on the incidence of homicidal violence are technical. In a perfect world of data collection, these obstacles could theoretically be overcome. However, other problems involving methodology may not be solvable simply by improved techniques. Reliance solely on law enforcement or legal classifications to assess homicidal behavior also proves insufficient. Part of the problem is the focus on arrest rates or conviction rates and the different degrees of homicide that can be charged (e.g., first-degree murder, manslaughter). These approaches are insufficient from a research perspective—and from the community interest perspective—because they ignore many of the similar types of behavior that may not result in a homicide. As I have often noted in this book, the difference between a homicide and an aggravated assault is often a matter of an inch’s variation in the path of a bullet. Hence, reliance on legal classifications for research is insufficient for understanding all the behavioral complexities of homicide. An expanded and reclassified approach would be needed, which would then raise a different set of problems. The nature and purpose of the research would determine whether such a reclassification, based on more than arrest and conviction rates and legal classifications, would be useful.

Homicide in the Twenty-First Century

395

Almost all professional disciplines are interested in developing research strategies that could provide a more valid and reliable assessment of homicidal behavior than currently available. Different techniques have been used to compensate for existing deficits.

Self-Report Studies One approach has been to use self-report studies, or victim surveys, in which reports are made about behaviors that could have resulted in a homicide or something close to it but did not. These events were never reported; why they were never officially reported, even as minor offenses, and how they might have been reported are questions that need to be addressed. Clearly perpetrators are not motivated to report such incidents. Nor in many cases do families or those with an intimate connection wish to report. Whether reporting homicidal levels of behavior should be encouraged or be made mandatory, as in alleged sexual abuses, raises complex moral and social issues. Another pervasive problem in studying homicidal violence from self-reports is that this type of violence has such a low base rate that it makes it seem impractical to use such methods. This handicap is specifically present when relying on surveys to seek out victims who could have been exposed to some level of homicidal violence.

Hospital Studies One variation of this research strategy has been to use hospital data on individuals who have sustained various types of serious injuries in the course of altercations. A major limitation of these data is their unreliability in terms of being an accurate estimate of homicidal types of injuries. One area in which this type of data gathering has been helpful is in investigations of the incidence of physical abuse of children from early infancy through childhood. The goal has been to obtain a more reliable and broader estimate of the degree and kind of serious physical assaults that are perpetrated on children. In the early stages of this approach, assaults were classified solely on a medical basis, and the type of physical injury received was the focus, without reference to etiology. This earlier system resembles the system used today for legal classifications of homicide. This hospital-based methodology could be extended to use data gathered from clinics and other health care facilities as well. In many ways this is a quite limited approach compared with the mandatory investigations into the nature of homicides that some European countries have. Violent deaths lead to turmoil and anguish for the

396


perpetrators and victims and raise community anger and guilt. In England, the National Health Service published “Executive Guidance” in 1993, which made inquiries into homicide inevitable for anyone who had any contact with mental health services. The process of investigation is referred to as “Inquiries After Homicide” and directs that in cases of homicide, it is always necessary to hold an inquiry that is independent of the providers who were involved.1

VICTIMOLOGY The role of victims of homicide was originally studied within the framework of victim-induced homicides. This approach was later criticized as one that blames the victim. However, in the context of homicides occurring between intimates, it must be acknowledged that systemic behaviors in the relationship can elicit homicidal behavior. Either party can often be seen as the perpetrator or the victim in the complex relationship that unfolds between two people, whether they be adults or an adult and a child. Victims’ roles in homicides need to be studied in greater depth than in initial studies. In those earlier studies, events were simply portrayed in terms of victim provocation, involving one key person who later had an act of violence perpetrated against him or her. Although homicides can occur within relationships characterized to some degree as masochistic, certainly that does not sufficiently explain the complexities and possibilities inherent in the dynamic relationships between people where an unusual outcome such as a homicide results. Another aspect of homicidal victimology research that requires more attention is clinical knowledge about the variety of interactions that may lead to violent outcomes between intimates. It is important to gain more knowledge about how some people seem to gravitate into a role of acting out violence in the complexities of their intimate personal relationships. If anything is known about relationships that have the potential to become violent, it is that neither party is a totally helpless individual in his or her interactions with others. Homicidal acts take place in the context of human interaction. It is also just as important to gain knowledge about why most people do not resort to violence in relationships, be they intimate or not. To address both of these questions, a multidisciplinary approach, rather than the segmented knowledge of one discipline, is crucial. The context of an intimate relationship is really only one model of close human interaction that can lead to a homicide. Similar dynamics


397

of victimology may operate in interactions with authority figures, such as larger or more physically powerful individuals, or those who are in authority over others, such as in workplace violence or in interactions in which people provoke others they see as exerting control over them (e.g., the police). After the histories of individuals involved in these situations were studied, a repetitious behavioral pattern appeared in that these individuals continually become involved in similar encounters that had the potential of turning into homicides, with these individuals as victims. It can be fruitful to reconstruct past consequences of their behavior and identify what factors increase those consequences to the level of these individuals’ being seriously injured or killed. Another area that needs investigation is situations with the potential for erupting in a homicide in which a certain individual has played alternating roles at different times. Clinical work appears to indicate that some individuals actually alternate between the roles of victim and perpetrator. One phenomenon that needs to be examined is these people’s past activities in which they have had violence directed against them, possibly by the person who later becomes their victim. Although retaliation may be relevant in some cases, it is usually more complex. These past activities often are found to involve high-risk or illegal behavior and include the individual threatening violence on others, such as threats of reprisal (either physical or social) or of extortion regarding another’s past violent actions. Threats to reputation or social stability operate here. There may be a subsequent pattern of continued provocation, with the potential for the situation to become destabilized and evolve into violence. However, these types of clinical impressions need empirical confirmation.

ROLE OF INSTRUMENTAL VIOLENCE IN HOMICIDAL VIOLENCE An operational perspective on homicide views such violence in terms of what is to be gained or achieved by the behavior. The gain in a homicidal type of assault is either the pleasure obtained in such an aggressive act or some type of retaliatory or self-protective mechanism. However, in some cases, such as a homicide taking place in the context of a robbery or rape, the explanation would be that killing was not the goal but simply instrumental to the main act. The initial goal was then simply a burglary, for example, but events got out of control so that the secondary violence of a homicide occurred. A derivative question then arises as to what key variables are operating for events that seem to have “accidental” homicidal endings.

398


An instrumental model of homicide may only provide a partial explanation, at best. More evidence may be required to explain these types of human interactions (i.e., the homicides) than viewing them merely as being instrumental in achieving another end. On the other hand, attributing more elaborate motives (e.g., killing as a “release” of hatred, killing as incidental to some other criminal act) to people who express violence in this manner may be an inaccurate attribution. Such attribution also leaves unanswered questions about whether people carrying out other types of offenses, in which violence is instrumental, are unaware not only of the role that is being attributed to them but also of the inferential processes for how a more complex set of motives is established, absent a type of intensive therapeutic investigation to unravel these complex questions. Some individuals would argue that the attribution of a role does not matter. Instead, they put forward the opinion that an individual’s degree of conscious awareness regarding key variables (e.g., the degree of risk he or she is taking, the degree of force used in perpetrating a certain act, how a particular victim is selected) does indeed matter. However, to get into such specificity about the cognitive operations and motives of violent offenders on an instrumental level goes far beyond the obvious explanation offered in terms of their behavior simply being seen as instrumental to a main goal (e.g., a burglary). Research into the complexities of motives is seen as worthy of pursuit. Thus, hypotheses need to go beyond the obvious operational explanation of the act serving some instrumental end. Whatever is learned about the instrumental aspects is then part of the overall story. Some researchers have suggested that instrumental models also may have application to studying the “legitimate” use of violence. Examples would be violence involving military personnel, violence in correctional facilities, violence in athletics, and violence exhibited by police officers. In fact, the apotheosis of retribution—capital punishment—can be viewed as an instrumentalized type of violence that is not only legitimized but approved by a majority of the American public, if we take as evidence that 38 American states and the federal government have had their elected representatives enact death penalty legislation.2 However, it is possible that such an inference is unwarranted. For example, in the United Kingdom, 88% of public opinion polls in 1994 were in favor of restoring the death penalty. However, 4 months later that year, the British House of Commons voted by a majority of 244 votes not to restore capital punishment for murder, and by a majority of 197 votes not to restore it even in the case of murdering a police officer.3 In many of these situations, there is the assumption that extreme violence is an inherent, legitimate part of military or civil law-and-order occupations. This le-


399

gitimacy model has also been applied to certain subcultural criminal groups, in which an analogy has been drawn between those groups and organizational groups in which violence is required to maintain discipline. The prime example is the consequences to someone who deviates within organized criminal groups by informing or cooperating with law enforcement agencies, such as in the Mafia. Even overzealous prosecutors may find their lives in danger.4,5 It would be valuable to study these supposed situations in which violence is legitimized to discern what conditions must be present in order for homicidal acts to occur. One relevant question is the extent to which the justification of instrumental violence is seen as inherent in a role that promotes the use of violence. Such violence may, in turn, elicit counterviolence by the potential victims. Another question to be answered is what predisposing personality variables influence people to choose roles in which they may ultimately have to resort to extreme violence.

SOCIAL MEANINGS OF VIOLENCE Apart from the legal consequences, what particular social or moral meaning may an act of homicide have? This question has been an important area for sociological investigation and philosophical speculation. The conceptual meaning of a homicide is often that determined simply by some type of official record keeping or the law enforcement process itself. However, the official meaning attributed to an act might be quite different from the meaning the act had for the perpetrator and society. To understand or explain such acts, it would be necessary to grasp the specific meaning of the homicidal act for the person performing it and how others in a particular social network react to it. Without this meaning, explanations are likely to be at least inadequate or partially invalid. Much research can be criticized for not seeking out the particular meaning of an act for the perpetrator. Worse yet, observers may rest content with the meaning given to homicidal acts by a particular agency that keeps official classifications, or by an agency whose job is social control. Too often the result is a set of stereotyped explanations that stifle further inquiry or offer rationalizations for the behavior in question. The positivist approach, in which behavior is interpreted as being determined by outside circumstances, poses grave limitations for researchers. However, research that accounts for the social meaning of particular acts would need to incorporate empirical as well as nonempirical approaches. For example, a preferred explanation could point out that an explanatory model might be viewed as causally adequate, yet have an insufficient level of meaning attached to it.

400


From a psychiatric viewpoint, considering the social meaning of an act is seen as necessary for research on extreme personal violence. In that sense, an examination of the social meaning of a homicide is an underlying necessity in any type of sociological research if it is to give cognizance to the psychiatric significance of the behavior. Joint psychiatric and sociological research on homicide is badly needed—in which the construction of the social meaning of homicidal acts is addressed, along with the way people attempt to cope and make sense of their world via such constructions. There is something unsatisfactory in theorizing about violent behavior without making reference to the individuals involved and what meaning a particular act may have had for them. Perceptions and motives that accompany the observed behavior are equally relevant. Although detailed ethnographic studies of violence have been conducted, what is needed is an attempt to forge a link between this type of work and more general theoretical constructs related to homicide. If that cannot be accomplished, the danger is that proposed explanations will be purely ad hoc or idiosyncratic. Theorists are aware of this dilemma, and they try to deal with it by resorting to approaches based on typing acts according to what meaning they seemed to have to the actor.

SOCIAL CONTROL AND VIOLENCE The legalization of physical coercion and actual violence, as when used by the state to control or punish people perceived as deviant, is a process that has continued throughout recorded history. However, the justification for such violence has never been sufficiently understood or adequately researched; this violence has simply been taken as a necessity. The relationship between social control and the use of extreme violence is one of the single most important areas that cry out for further explanation. Social control is a very broad concept. Although part of this question has been examined by sociologists, further investigation is needed, such as how individuals react when coercive power is exerted on them. The goal of such research is to elaborate on the distinction between what is described as legitimate force used to control others versus the force connected with criminal violence. The questions that should be addressed are whether there is a legitimate basis for the distinction and, if so, what validity such a distinction has. These questions are particularly germane to explanations put forward by state agencies that carry out social control via use of force, such as when those authorities sanction a subgroup to carry out violence; for example, these


401

groups could be set up for the purpose of regulating illicit drug distribution, setting up security units for those classified as dangerous in prisons, or acting as private security forces. The effects these approaches have on increasing the possibility of extreme violence by both the controllers and the controlled need careful investigation. Relevant to this issue is how organizational changes may affect the perception of the enforcing agents in terms of their feeling a need to use increased force to achieve their mission. Different people may carry out these missions quite differently. A changed conceptualization of their role may mean that they start to perceive the nature of their interactions in such a way as to increase the probability of a homicidal act. If death occurs, it is more likely that those who see themselves as the recipients of such social control may strike out as a backlash because they perceive themselves to be potential victims of violence. In this context, there is a need for more knowledge about the sociology of law, such as how law is used in society, and to integrate this knowledge with a psychiatric understanding of the dynamics of provocation. This integration would encompass investigating the legality of criminal violence carried out by agents of social control, how they perceive their mission, the origins of law about the use of violence, and how the processes of enactment of these laws influence their later application. Psychiatric input in these processes may well contribute to better understanding of confrontational types of violence and homicide. The last third of the twentieth century witnessed a shift in American crime control policies regarding serious violence. Punishment became harsher and long prison sentences normative. Capital punishment became a possibility for homicide in 38 states and in federal jurisdictions. Many reasons have been suggested as explanations, such as the severity of the punishment simply paralleling the increase in homicide rates. It was often stated that the public demanded increased punishments. Such demand seemed to reflect a breakdown in the idea that the government could be relied on to control violence unless such harsh measures were employed. The belief that the government could not provide routine security against serious violence promoted a turn toward severe punishments as a logical outcome. Politicians were likely to see the situation as one in which they could easily capitalize by demanding harsher penalties. Unfortunately, such a simple solution bypassed approaches of trying to understand the origins of homicidal behavior and to seek relevant preventive measures. At best, a question-begging explanation was offered that socioeconomic changes lay behind the increased rates of homicide that had emerged by the early 1990s. This explanation persisted even

402


though the rates had dropped by the end of the decade and subsequently. The approaches adopted remain in effect. The focus was more on immediate events rather on than taking a long-term viewpoint that cycles of violence occur historically. The sociologist David Garland6 described a “culture of control” in his expanded explanation of societal changes and violent behavior. The picture is that from the 1970s, high rates of crime with an accompanying fear of crime became a fact in American society. Avoiding crime became a prominent principle of life not only in the lower socioeconomic strata but in the upper classes as well. Increased homicide rates led to uncertainty that the government could control such behavior. In turn, this uncertainty elicited in government and political agencies the feeling that they had to do something. The solution was to move in the direction of increased penal sanctions to convey that they could actually do something. The effort was to try to reassure those in positions of power and the ruling classes that certain groups were to be regarded as dangerous (“the criminology of the other”) and that by controlling these groups the public would be reassured that something was being done and violence was not going to be tolerated. In turn, Tonry7 thought that this explanation, although cogent and sophisticated, did not suffice to explain the milieu of harsh punishment in which the U.S. imprisonment rate is nearly five times higher than in any other Western country, even ignoring capital punishment. Gross crime trends were seen as determined by fundamental social and structural forces in a society. Consequently, short-term changes in penal practices would not be seen as affecting rates or patterns of crime except at the margins. Rather, short-term panics occur where problems become exaggerated, and extreme policies are the result. Tonry’s position is that the misfortune of our time has occurred because long-term deviance cycles, accompanied by intolerance and excessive severity, have coincided with a series of short-term moral panics. The short-term panics then exacerbate the effects of the long-term cycle. This critique is consistent with the original work of the nineteenthcentury French sociologist Emile Durkheim.8 Durkheim held that one of the main functions of the criminal law was to reinforce basic social ideas about right and wrong. Hence the criminal law enacted a drama in which punishment was primarily for the sake of the community and not the offender. Punishments worked via affirmation of prevailing norms and not by seeking rehabilitation, deterrence, or incapacitation per se. The goal was to maintain social cohesion and only secondarily control crime rates or deal with why offenders behaved the way they did. Legal approaches were thus to serve as a background to socializing


403

the values of institutions. Cycles and panics play a powerful role in legal approaches by influencing the sensibilities of a society at any particular time.

GROUP VIOLENCE LEADING TO HOMICIDE Attempts are often made to distinguish criminally motivated violent behavior from politically motivated violent behavior. It has also been argued that such a distinction is not totally valid. This argument does not mean that we should attribute a conscious intention to all so-called political activities that become violent. The fundamental position is that politics is one way of dealing with social conflicts in an effort to produce order. However, similar activities to secure such ends may also be criminal. One implication is that violent behavior cannot always be assumed to be irrational. All too often such an assumption is made when group violence occurs, particularly when accompanied by a homicide. Yet violent acts that occur in a group setting often have great political or moral meaning for people. Whether the killings between different ethnic and cultural groups are rational acts of patriotism designed to achieve legitimate ends or rather behavior that is closer to that engaged in by the criminal or insane mind is a debatable question. Why religious militants kill is a question in its own right. Their behavior may be motivated by a mixture of spiritual and political goals but still have components of mental disturbance.9 Perhaps that is why it is so difficult to understand, let alone stop, the killings that have occurred in the areas of Croatia, Bosnia, and Serbia and subsequent types of terroristic killings, such as those in Iraq. Some of these acts may be carried out based on long-smoldering hatred and needs for revenge that are important to understand—but these acts may also be committed because of reasons more closely connected to psychopathology in some individual perpetrators. On a more specific level, homicides connected with assassinations or terrorism need psychiatric investigation and research in their own right to gain some additional understanding. Some violent situations can be described as “mob violence” that escalates and results in assaults and killings. The psychology of the mob is studied in the context of what different psychological and social factors coalesce with individual dynamics. The periodic eruptions of these phenomena—in such diverse settings as sporting events, political rallies, and gang killings—are not understood in depth, even though they are often viewed as normative under the circumstances. Techniques for preventing group confrontations have worked only minimally. A related issue is

404


some groups’ social expectations that the way to solve problems is by direct confrontation and the avoidance of any type of compromise. Another example of group violence is the slaughter of civilians carried out by military personnel. For some people, these acts of mass violence may be seen as instrumental to achieving a particular goal (e.g., fostering changes in society, punishing opponents, expressing deep feelings of hatred), and the perpetrators may be perfectly willing to pay the price that accompanies the commission of the crime. However, some of these individuals may also have varying degrees of psychopathology; it would be important to determine whether certain individuals who were carrying out the same act within the same agenda actually had quite different types of conflicts or personality traits. In this context, inquiry is needed about the social and political processes by which certain fostered behavior is defined as violent. The motives and consequences of using different definitions of violent acts are part of this effort. Youth gangs, whose numbers ebb and flow in major American cities, may claim they are acting to foster a political agenda. Although this motivation may be true for some, for others it is a rationalization for homicidal violence. Those who perpetrate mass killings or serial killings similarly need investigation on various levels (see discussion in Chapter 1, “Epidemiological Aspects of Homicide”).

VIOLENT INDIVIDUALS Predisposed Individuals A traditional and comfortable approach for psychiatrists researching homicidal violence is to focus on the individual perpetrators who have committed acts of violence. The focus may take either a neuropsychiatric or psychopathological direction in the search for factors that contribute to the personality structure of an individual. Apart from cases in which an insanity defense is raised with specific legal questions, the goal of such research is often to see what makes the person potentially vulnerable to commit homicidal types of violence. However, there may be an implicit assumption that previously violent individuals are the ones who later commit homicides, although this hypothesis has not been justified. Some individuals merely continue being violent in diverse ways, with a diminution in their violent behavior as they age. Researchers who focus on social and environmental factors that contribute to violence often feel that clinicians overemphasize individual factors. The opposite type of criticism comes from those who feel that individual psychodynamic factors are too often ignored by those who


405

compile sociological data. Many of these factors have been discussed in different chapters of this book and are simply noted here. Researchers of homicidal individuals must necessarily acknowledge the diverse variables involved in those individuals’ behavior. Some researchers have dealt with people’s innate aggressiveness in an attempt to understand biological variables as a condition for the expression of violent aggressive impulses. Research in developmental psychopathology has focused on the outcomes of different child-rearing practices and aggressive behavior. However, to a great degree, the latter type of research has comprised more studies on minor types of aggressive displays than studies in which certain behaviors and traits are examined for predictive significance for a person’s later proneness to violent behavior. There is also the recurrent problem of the different socioeconomic statuses of families and what impact this has on types and kinds of aggressive behavior.

Role of Punishment The particular kind of punishment used in families is one variable that has been studied a good deal. Unfortunately, many of the studies comparing physical punishment with educational techniques as means of controlling behavior have also focused on more minor types of overt aggression, such as behavior in classrooms, in contrast to serious assaultive violence. This latter outcome needs to be studied to understand the relationship of punishment techniques to later homicidal types of behavior. The influence of disciplinary techniques with respect to the later occurrence of extreme violence is a very complex question. Long-range projections are also often fraught with low validity. A significant difference may exist between the conditions operating in experimental laboratories, in which punishment is administered in the form of mild electric shocks to humans or nonhuman species, and the ways discipline is administered by parental figures. Making any outcome predictions about a later homicide is quite precarious because of the multiplicity of variables that operate in homicidal situations and the relatively rare occurrence of homicide. The factor of living in a culture in which physical violence frequently occurs or is witnessed—directly or indirectly—cannot be ignored as having some influence on vulnerable people. The impact of the mass media and movies is now an inherent part of growing up in America. Often the histories of criminally violent juvenile and youthful offenders who eventually commit a homicidal act reveal that these individuals were exposed to violence early in life. However, it would be precarious to

406


suggest that the causal effects should be interpreted as linear. Much more specific longitudinal studies are needed to delineate the possible effects of these variables on an individual who later engages in homicidal behavior. Experiencing violence is also different than witnessing it. If experiments were run exposing children to witnessing violence in a laboratory setting, assuming these experiments could be done ethically, many questions would still remain open about the long-range effects of such viewing, particularly with respect to future homicides. Given the amount of serious violence in the United States, both experienced and witnessed, a safe conclusion seems to be that most of those people exposed do not become that violent themselves—if they did, even more homicides would occur.

Physiological Variations In discussions of innate differences among individuals, the factor of conditioning that has promoted the expression of violence is often raised as a contributory variable. Arguments about the relative importance of such learning experiences versus the significance of physiological aberrations in the autonomic nervous system are currently impossible to resolve, primarily because these two kinds of variables interact to such a degree.10 For example, psychopathic individuals, or those clinically classified as having antisocial personality disorder, have been studied for different autonomic reactions, such as their lowered responses to certain kinds of stress or the difficulties in conditioning these individuals for avoidance reactions. Because serotonergic neurons heavily innervate frontal temporal regions, which are implicated in psychopathy and impulsive aggression, it is possible that some type of neurodevelopmental aberration may be contributing to altered processing of affective stimuli in vulnerable cortical regions.11 Along biological lines, the entire realm of genetic differences will play a prominent role in future research with this group. Key antecedent variables in assessing a proneness to violence and that likely will have a high research payoff are 1) the possibility of seizure disorders or cerebral types of injuries that occur either prenatally or during the birth process; and 2) infectious and traumatic processes. Early recognition of those persons with neonatal brain injuries by advanced methods of neuroimaging may play a key role in reducing the later incidence of seizures and behavioral disturbances.12 The theoretical framework is that such types of insults to young, developing humans predispose them to later difficulties in the control of aggression or


407

to abnormal rage reactions of the episodic dyscontrol variety. The cyclicity of anger and rage in individuals with borderline personality disorder is just beginning to be understood in terms of these people’s inherent instabilities, provocativeness, and explosiveness.

Ego Functioning As noted earlier, the basic approaches to the study of individuals who have committed some type of violent act are to focus on either their biological predispositions or preexisting psychodynamic conflicts. One hopes that this type of oversimplified dichotomy has had its day, for both approaches need to incorporate significant social variables in their explanatory models along with a study of ego functioning. Criticism of both approaches has largely been on methodological grounds because of difficulties in generalizing from one individual to another in ideographic approaches. Nevertheless, the case history, or clinical assessment, approach can reveal immense details through gathering data on the developmental history and lifestyle of someone who has committed a homicidal act. The case history approach may reveal important information—for example, the types of intrapsychic conflicts operating and the psychological defenses that have failed, thus allowing the violent eruption of homicidal acts. What contributes to the failure of defenses in such circumstances also needs more detailed investigation. The study of the failure of certain defenses with different types of killings, as correlated with different diagnoses, may prove quite fruitful. In those individuals prone to homicidal violence, the role of conscience development, superego failures, and deviations or failures in socialization need examination.

Homicide Without an Explanation More bizarre types of killings, sometimes referred to as those without an explanation, call out for intense psychiatric scrutiny. These homicides would include those that occur in the context of sexual encounters, random assaultive acts on unknown individuals that result in homicide, serial killings, mass acts of violence, and homicidal violence perpetrated on or by children. These unfortunate occurrences need to be studied to acquire knowledge about the propensities for such an outcome if these homicides are to be better contained in the future. A delusional thought process that has been quiescent but then becomes activated in terms of taking a homicidal course merits investigation: what factors finally tip

408


the scales in these people compared with what allows the majority of people with such delusional beliefs not to take action? Cases also merit research in which an isolated and extreme violent act is committed by a person who heretofore had not shown any violent tendencies. A variation is the analogous situation where an eruption of violence occurs under changed life circumstances. The most common is the person in a state of unresolved mourning who commits a homicide. Those persons who are in a state of severe depression are usually studied as suicidal possibilities, with homicide being ignored; this is a largely overlooked group in which these variables are missed. Under military conditions, some soldiers may go beyond the norms of combat and begin to torture or kill prisoners or civilians. A more refined version of this killing occurs when previously “normal,” or at least nonviolent, individuals are working in a milieu, such as a prison or concentration camp, where they conform and adapt to the killing circumstances. Perhaps killings that occur in the context of being a guard or police officer are similar in that sanctions officially allow violence. Historical examples might be those individuals who have functioned as hangmen or executioners. There is also the larger group of persons who join in the activities as onlookers and receive some type of vicarious enjoyment from the activities, or those individuals who silently acquiesce as passive participants, providing in effect a silent proxy. Related are the current onslaughts of killing in Africa, Asia, and Latin America from nationalist movements, ethnic rivalries, insurgent political groups, and regional conflicts, in which civilians are the most frequent casualties.

Intoxicated States Drug use and alcohol intoxication are well-known contributors to severe violence. However, why they operate more prominently to induce violent acts in some people but not others is the question. The usual explanation given for the effects of drugs and alcohol is that they facilitate violence by lowering the threshold for responding to provocation. Yet this explanation is more like stating the obvious. The acts may occur in a social milieu in which defenses have already been lowered. Indications are that explanations may not be that straightforward; for example, subtleties exist in terms of the effects of chronic drug and alcohol usage, and violent behavior can accompany withdrawal syndromes when usage ceases. Diverse factors need to be addressed, such as individual predispositions to the effects of alcohol or drugs and the longterm effects of different combinations of drugs and alcohol that can contribute to violent acts. In essence, the role of drugs or alcohol as a con-


409

tributory factor to a homicide involves a complex interaction between the pharmacological properties of the substance, the different individual reactions, and the precise social circumstances in which the act occurs. In addition, the presence of a comorbid disorder is always a possible complicating factor.

VIOLENT SITUATIONS Escalating Encounters Considering situations shifts the focus to specific types of interactions that take place between individuals that eventually result in a homicide, rather than the behavior of individuals themselves (see discussion in Chapter 7, “Masochism and Homicide,” on masochistic phenomena and shifting victim roles). Tense interactions in which the potential for violence exists may occur between individuals over a prolonged period of time or may emerge from a combination of intense but brief interactions. When looked at more closely, these situations often can be assessed in terms of interactional processes that escalate because of slight miscues and misinterpretations that then begin to be perceived as serious threats. The aggressive responses become heightened on both sides, further retreats are made, and the probability of reaching a point of no return increases. If the tension does not subside quickly, an explosive and violent outcome becomes likely. A better grasp of the details of these interactions and escalating variables is needed, for they are the settings in which increased prevention will be possible for one large group of homicides. A process analogous to the “psychological autopsy” performed after a suicide can be fruitful in attempting to reconstruct what happened and identify the key variables. It is hoped that more knowledge of these processes can lead to better intervention by people who either are on site or are called on to intervene in these situations. Not only the police but also individuals who work in clinical settings or in violence-prevalent areas need to learn ways to diffuse such escalating encounters. Perhaps bartenders and “bouncers” will become the key preventive interveners, given that escalating scenes frequently occur in a drinking atmosphere, just as police are key people in the ubiquitous escalating domestic disputes.

Mass Media The role of the mass media in heightening the potential for violence is similar to the impact of propaganda or advertising. The media are one contributory variable in a culture that already emphasizes such things as

410


the “macho” way (e.g., through confrontation, reliance on weapons) to settle disputes. All too often the media present this situation in a dramatic format, and the subsequent relevance of modeling influences on the watchers is disturbing. However, it is difficult to confirm or disconfirm logically whether the mass media are one of the main culprits in contributing to violence when similar acts of violence occur that seem to mimic the violence seen on television or in the movies. Conversely, can the media be part of a solution in curbing violence? What occurs in watchers is a complex interaction that varies from one case to the next. The massive exposure to such modeling of aggression via television and movies, which individuals receive from a young age onward, needs assessment. Even more difficult to unravel are the effects of pornography. There are those who ardently advocate that seeing pornography contributes to violence and sexual murders. Of course, there are equally ardent critics on the other side of the argument. There is also the phenomenon of individuals being willing to administer pain to helpless subjects, as has been demonstrated in social experiments using severe electric shocks. The conclusion is that individuals will conform to whatever environment is seen as acceptable at a particular time, especially when such behavior is sanctioned by authority figures. Further analysis is required to determine what individual variables cause some people to succumb in such circumstances. One interesting side question is whether there would be increasing participation by females in violent confrontations of this mode, given the blurring of boundaries in social roles of males and females. One hypothesis—which is perhaps more of a hope—is that males may not need to continue to rely on violence as a way of asserting their masculinity to the degree they have in the past because of the social changes in male and female roles that have taken place. If a decrease in violence does occur, one hypothesis is that it will be attributable to the increasing constraints placed on males to avoid violence against females. There is also the opposite possibility: that constraints on the use of physical violence by females will be lessened because that behavior has become more permissible for them, a paradoxical effect of equality. On a larger scale, the idea that more men and women need to arm themselves may override all these arguments in promoting more homicides.

PREDISPOSITION OF HOMICIDAL INDIVIDUALS WITHIN THE LEGAL SYSTEM Many unresolved questions remain among the usual problems and legal issues that arise in dealing with homicidally violent people in courts.


411

Capital Punishment One issue is the effect of the increased prevalence of capital punishment statutes and the increase of those individuals sentenced under those statutes—along with the effect that capital punishment will have on all members of society. What effect capital punishment will have in terms of aggravation or mitigation of violent behavior in the future remains largely speculative. Although ever-larger numbers of individuals continue to be convicted and to take their place on death row to wait through the years of the appellate process, the debate on the use and effectiveness of capital punishment from economic, moral, and deterrence points of view continues. An exception involved Daryl Atkins, a man with an IQ of 59. In Atkins v. Virginia,13 the U.S. Supreme Court decided in 2002 that mentally retarded individuals are less culpable and therefore cannot be executed. The reasoning was based on the position that a consensus of states did not permit it. Use of capital punishment seems to go in cycles. Diverse explanations have been offered to account for the increased use of capital punishment and the accumulation of people on death row—from the conservative shift in American politics with an emphasis on states’ rights to psychiatric justifications that the relatives of homicide victims can achieve “closure” by executions. Zimring14 pointed out that the states that execute are mainly the old slave states. In 2000, 76 of the 85 executions (89%) were in the South, even though that region accounts for only one-third of the U.S. population. In that same year, two-thirds of U.S. executions were conducted in just 3 of the 38 states that authorize capital punishment (Texas, Oklahoma, and Virginia). Because of the increased reliance on capital punishment, the usual forensic questions that are customarily raised when capital punishment is a possibility, such as insanity defenses and competency issues, will undoubtedly continue to be raised and even be expanded. One result of capital punishment is that defendants in those jurisdictions are forced to fall back on the most unreliable and least used approaches, such as insanity. Ethical issues also arise after convictions when psychiatrists are asked to treat individuals ruled incompetent for execution so that these individuals can then be executed. Perhaps one of the least appreciated consequences of capital punishment for first-degree murder convictions is that communities are lulled into thinking they have adopted a solution to control homicidal violence; instead, the case is simply that states are exacting retribution for increased homicidal violence on a few convicted killers.

412


Sentencing Trends Similar questions arise about the possible effect of different sentencing practices on homicidal behavior, such as the increased frequency of incarceration for minor offenses and the increased length of sentences for most offenses. Real-offense sentencing is another practice under federal sentencing guidelines where sentences are calculated by a judge on the basis of not only the convicted offense but also offenses for which the person was not charged or for which the charges were dismissed or dropped. The hypothesis is that these practices remove violent individuals from society and hence may lower the homicide rate. A contrary hypothesis is that increasingly punitive sanctions foster the use of violence—either within the prison setting or within the civilian community, where individuals carry back the experiences of prison violence after they are released. Some hypotheses offer an analogy between hardened criminals and a subset of veterans who return from combat in which they were exposed to violence and heavy drug use under certain conditions and who continue on such a pathway. Although these phenomena pertain to a minority of individuals, they raise questions about the need to examine crucial factors continually and come up with better explanations for why some individuals are homicidal, as well as why communities attempting to control serious acts of violence are experiencing a sense of hopelessness.

Relationship of Other Offenses to Homicide There is one criminological question where psychiatry may be able to make a critical supplemental contribution. The question concerns the relationship of other types of offenses (e.g., property offenses) to homicidal offenses and what factors push an individual to become increasingly violent, if not homicidal, over time. The study of criminal careers in this context is the key to obtaining a better understanding of this group. One type of property offender may have an only occasional violent eruption, whereas others seem to progress to the seemingly inevitable end-point of homicide; the majority may confine themselves to property offenses. A study of reconviction rates and risks can focus on the assessment of progressive types of dangerousness. Prediction tables have been constructed in relation to developing sentencing grids; developers of these tables have tried to rely on objective data such as previous convictions, age, marital status, socioeconomic class, and employment records for the purpose of determining


413

the crucial variables in determining the severity of the punishment. Unfortunately, in these grids, little weight is given to individual psychological factors. The question is whether explanations about such shifts in behavior can be better predicted if individual personality measures with clinical observations are also considered. Very little is known about the effect of treatment interventions on extremely violent individuals and whether these interventions elicit a changed outcome. This is partly because these individuals rarely receive such interventions, and for the few who do, there have been no prolonged and controlled investigations to provide reliable conclusions. The types and durations of treatment need specification, as do the theoretical basis for the treatment and the characteristics of those carrying out the treatment, before any conclusions about outcome are warranted.

Diagnostic Precision Psychiatric contributions to understanding homicide are predicated on the validity of diagnostic classifications. Although reliability studies have multiplied, as well as the number of diagnoses with each edition of the diagnostic manuals, questions about the validity of the diagnostic entities remain. Despite attempts to use empirical testing for inclusion, acceptance of the diagnostic formulations rests with committees and advocates. Inquiries about the validity of diagnostic frameworks are unfortunately often confused with philosophical conjectures. The latter ask whether diagnostic entities have a “real existence”—meaning, do they correspond to something that exists in nature? Put another way, do the diagnostic groups—to which those working in psychopathology give names, such as major depressive disorder—describe a set of signs and symptoms that exist even if never observed and given such a name by a psychiatrist? A scientific realist says yes, whereas a scientific fictionalist says no. However, this type of epistemological question should not be confused with the scientific question of how the signs and symptoms should be classified. This is basically an empirical problem to be solved ultimately by applied mathematics.15 In the future, adopting this more rigorous approach to diagnostic classification should allow substantive gains to be made in understanding certain homicides. By virtue of greater diagnostic accuracy, more specificity will occur to connect homicidal behavior with diagnostic categories. In the meantime, lacking a gold standard, the best we can do is to use a “boot strap” system of classification to imply the possibility of real classes by way of presumed indicators.

414


SECONDARY EFFECTS OF HOMICIDE Another area that merits increased attention concerns the ramifications of a homicide. The sequelae extend far beyond the individual case processed in the criminal justice system and the type of judgment or verdict given. Understanding some of the psychopathology connected to such violence in the individual also does not exhaust the significance of the acts in a society. The offender’s family is left to deal with his or her absence for an extended period of time. A good number of marriages end during that period. The emotional effects on children are not only disruptions in attachments and loss but also include the ridicule, teasing, and ostracism that sometimes occur from a peer group. Similar processes operate with the families and loved ones of the victims. Estimates are that each homicide victim is survived by an average of three loved ones who must deal with the homicide.16 Amazingly, the children of a victim may also bear the burden from a peer group that abandons them. Thus, variations in the bereavement processes of “secondary victims” occur after a violent death. The impact induces a state of emptiness and deprivation. The victims’ families must deal with an unrequested loss, with accompanying feelings of sadness. A number of dramatic responses occur after the initial fear and disbelief. These responses often produce symptoms of anxiety and depression, with somatic accompaniments such as difficulty concentrating, emotional instability, startle responses, and sleep difficulties; it is not difficult in many of these situations to see signs and symptoms of acute stress disorder, if not full-blown posttraumatic stress disorder.17 Loss of a loved one by a homicide is complicated by several factors, such as its unexpectedness as well as the prolonged anger and rage that may be fed by extended legal maneuverings. When some type of legal resolution is finally attained, dissipation of the anger does not necessarily occur. Ruminations may persist, with fantasies of revenge and retribution. Family members may affiliate with organizations dedicated to those who have had a murder occur in their families. Some reality problems occur immediately, such as needing to make funeral arrangements, writing obituary notices, dealing with any media who seek interviews with family members, and so on. However, there are also prolonged reality problems, such as maintaining the family income, making custody arrangements of children, and settling inheritance issues. All 50 states have crime victim compensation awards but these primarily relate to direct expenditures for funerals and payments only if losses cannot be recouped from other sources.


415

Some homicides leave special bereavement problems. One example is the homicide for which the killer is never found, making it difficult for victims’ family and friends to attain closure. If a perpetrator is charged years later, all the initial reactions may reoccur. The police reopening a case years later on a new lead is always a possibility. Periodic media coverage may also cause complications in bereavement. Lurking for years after the homicide is always the question of forgiveness, mixed with the difficulties of efforts to forget. Generalizations can be offered from empirical work on who has the most difficulty in coping with a homicidal loss.18 Those in the lower socioeconomic classes may have more difficulties because they have fewer resources available to them. Men are seen as having more problems compared with women because they do not express their feelings as readily and have fewer networks available to them. Younger people may be dealing with the first significant loss in their lives, which may seem overwhelming. It may not only be the suddenness of the loss that has an effect. If brutal aspects have been present, such as a child being sexually assaulted and then murdered, or if there was evidence of torture, an added burden is present. Violent deaths in children are particularly difficult. Conflicts with the deceased that have not been resolved are an added burden to bereavement processes. An intrafamilial killing has the added dimension of the murderer being someone to whom there is still a deep attachment. In all of these situations, the burden is not only the occurrence of a homicide but also how the homicide is experienced by those who are touched by it. Those who observe a homicide respond in their own way. Children who have witnessed the murder of a parent have an additional burden beyond the loss of the parent.19 Throughout all of these situations two discrepant perspectives operate. One is the approach to homicide with a sterile set of statistical and demographic data—who does what to whom. This scholarly approach reflects how academics and experts discuss types of homicides. A contrasting viewpoint is the experiential one of “secondary victims” who demand recognition and participation. Their situation is conveyed as a devastating symbolic passage that is not totally intelligible to outsiders who have not had such a profound personal experience.20 Both of these perspectives will need added sophistication in the twenty-first century for better understanding of these complications and to limit secondary emotional complications.

416


CONCLUSION Each chapter in this volume contains the seeds for research sufficient to last well into the twenty-first century. What becomes more evident in studying homicidal people is the strain between those who believe that the emphasis should be on a greater academic sophistication in appraising the problem versus those who want more clinically attuned and forensically relevant output. In this book, I have revealed how diverse topics overlap and actually continue to expand by virtue of the increased specificity that has occurred in psychiatric diagnostic processes. The overall concern with violence bodes well for acceptance of the need for deeper research. At the same time, this concern is a sad commentary on the state of violence in many parts of the United States. Besides the contributions psychiatry can make toward improved validity and reliability in the diagnoses of homicidal individuals, it is hoped that psychiatry can also help researchers to achieve greater specificity in matching diverse kinds of homicides with specific diagnostic groupings. The problem of risk assessment for potential homicidal behavior continues to loom as large as ever. Increased knowledge from biological aspects of aggression will undoubtedly help us understand some aspects of homicidal behavior. One of the confusing elements in studying such violence is how much of the past work has focused not on homicide per se but rather on studies of aggression or at best on generalized definitions of violence. Such generalized studies on aggression and violence cover an enormous breadth of behaviors. They are all interesting in their own right, yet not all are germane to the problem of homicide. The result is a seemingly endless list of articles and books dealing with these topics. Sometimes the emphasis is biological, whereas at other times it is psychosocial, with an emphasis on social learning theory, the role of attachment, and similar factors. Yet other researchers believe the focus should be rather on sociological variables, such as the violence that continues to flourish among the underclasses. In the milieu of the inner cities in the United States, deterioration continues in the midst of a culture of poverty, drug use, and drug distribution. Accompanying this subculture is the spreading incidence of juvenile homicides among younger and younger individuals, as both perpetrators and victims. The prevalence of handguns and hours of exposure to television violence that children receive during their developmental years will merit increased scrutiny, because we will likely see the continuation of extreme forms of juvenile violence. Given the societal emphasis seen in past work on homicide, a critic may wonder whether there is a place for a psychiatric perspective on ho-


417

micide. The orientation of this volume has been that regardless of what social hypotheses are offered, these suppositions do not bypass the need for correlating the acts of individuals with careful psychiatric assessment. If anything, the lack of such psychiatric work in the past has been a major deficiency in studies on homicide and violence. Whatever changes occur in the social milieu of people—whether demographic changes, economic shifts, or governmental policy modifications—an individual actor always ultimately perpetrates a killing. A psychiatric perspective will continue to focus on these individual actors, much as will the legal system out of its need to assess blameworthiness. Although the prevalence of violence among intimates has declined relative to violence perpetrated against strangers, violence among intimates still accounts for a majority of the homicides occurring in the United States. The variety of biological predispositions and psychodynamic variables operating in the area of violence between intimates leading to homicides is much more complex than originally thought. This type of violence was once conceptualized around a few discrete situations: the occurrence of delusions of infidelity leading to a spousal killing, or a wife striking back at her husband after a long history of physical abuse. The context of such violence has now been widened considerably—for example, to children killing parents they allege have physically or sexually abused them, and battered wives who, because of the fear engendered in them by their spouse, strike out first and kill their spouse in the belief that this killing is necessary to defend themselves against a future homicide attack. Questions about homicides in response to incest, marital rape, husband battering, abuse of the elderly, familicides, and the frequent role of drugs and alcohol in homicides all speak to the need for careful psychiatric appraisal in addition to an awareness of the social milieu in which homicides occur. The impact of homicides on the families and friends of victims as well as on those of the perpetrator merits much more attention than it has received. Clearly, clinical fields such as psychiatry will continue to play a significant role in this important area of human behavior throughout the twenty-first century.

REFERENCES 1. Peay J: Themes and questions: the inquiry in context, in Inquiries After Homicide. Edited by Peay J. London, England, Duckworth, 1996, pp 9–38 2. Law enforcement and crime, in Information Please: Almanac Atlas and Yearbook 1995, 48th Edition. Edited by Johnson O. New York, Houghton Mifflin, 1995, p 859

418


3. Rutherford A: Crime, law enforcement, and penology, in Britannica Book of the Year 1995. Chicago, IL, Encyclopedia Britannica, 1995, pp 145–148 4. Jacobs JB: Busting the Mob: United States v Cosa Nostra. New York, New York University Press, 1994 5. Stille A: Excellent Cadavers: The Mafia and the Death of the First Italian Republic. New York, Pantheon, 1995 6. Garland D: The Culture of Control: Crime and Social Order in Contemporary Society. Chicago, IL, University of Chicago Press, 2001 7. Tonry M: Thinking About Crime: Sense and Sensibility in American Penal Culture. New York, Oxford University Press, 2004 8. Durkheim E: The Division of Labor in Society (1893). Translated by George Simpson. New York, Free Press, 1933 9. Stern J: Terror in the Name of God: Why Religious Militants Kill. New York, HarperCollins, 2003 10. Volavka J: Neurobiology of Violence, 2nd Edition. Washington, DC, American Psychiatric Association, 2002 11. Siever LJ: Neurobiology in psychopathy, in Psychopathy: Antisocial, Criminal, and Violent Behavior. Edited by Millon T, Simonsen E, Birket-Smith M, et al. New York, Guilford, 1998, pp 231–246 12. Ferriero DM: Neonatal brain injury. N Engl J Med 351:1985–1995, 2004 13. Atkins v Virginia, 122 S Ct 2242 (2002) 14. Zimring BF: The Contradictions of American Capital Punishment. New York, Oxford University Press, 2003 15. Meehl PE: Bootstraps taxometrics: solving the classification problem in psychopathology. Am Psychol 50:266–275, 1995 16. Alvarez A, Bachman R: Murder American Style. Belmont, CA, Wadsworth/Thomson Learning, 2003 17. Rynearson EK, McCreery JM: Bereavement after homicide: a synergism of trauma and loss. Am J Psychiatry 150:258–261, 1993 18. Rock P: After Homicide. New York, Oxford University Press, 1998 19. Malmquist CP: Children who witness parental murder: posttraumatic aspects. J Am Acad Child Psychiatry 25:320–325, 1986 20. Rock P: Murderers, victims and “survivors.” Br J Criminol 38:185–200, 1998

INDEX Page numbers printed in boldface type refer to tables or figures.

Abandonment, and borderline personality disorder, 141–143 Abortion, and decline in homicide rates, 49 Accomplices, to homicide, 308–309 Action-reaction relationship, and substance abuse, 24 Action systems theory, on schoolrelated homicides, 320 Actus reus (criminal act), 69 Act versus actor, in legal and clinical views of homicide, 368–369 Acute catathymic homicides, 352 Addiction, and dependent personality disorder, 158. See also Substance abuse ADHD (attention-deficit/ hyperactivity disorder), and juvenile homicide, 293, 295, 296, 312 Adjustment disorder, 385 Adolescents. See also Age; Juveniles antidepressants and suicides in, 272–273, 274 gunshot wounds as cause of death in, 284 homicide and suicide as causes of death in, 286, 309 sexual homicide and, 347 Adoption studies, of violent behavior, 60

Advocacy, and battered woman syndrome, 211–212 Affective instability, and borderline personality disorder, 125, 133– 134, 135–137 Age, and epidemiology of homicide, 3–4, 5–8, 19, 20–21. See also Children; Infants and infanticide; Juveniles Aggravated assault, 49, 394 Aggression American culture and, 28 controversy on use of term, 55 dependent personality disorder and, 171, 174 early theories on sources of, 56–59 formal functions and acceptable levels of, 373 seizure disorders and, 65 Agitation, and psychotic depression, 254, 257 Alcohol abuse and alcoholism. See Substance abuse Alien self, and borderline personality disorder, 130 Ambivalence, and dependent personality disorder, 164 American Law Institute, 368 Amitriptyline, 271 Amnesia, and dissociative disorders, 314–315

419

420 Amphetamines, 25 Anger. See also Rage borderline personality disorder and, 123, 134, 135, 136 dependent personality disorder and, 165, 171 depression and attacks of, 242 secondary effects of homicide and, 414 Animal models, of aggression, 55–56, 57, 58 Antidepressants. See Selective serotonin reuptake inhibitors (SSRIs) Antisocial personality disorder borderline personality disorder and, 126 juvenile homicide and, 293–299, 300 narcissistic personality disorder and, 178 neurophysiology and, 406 Anxiety, and dependent personality disorder, 167 Arrest rates, and studies of schizophrenia and violence, 99– 101. See also Law enforcement; Recidivism Arson juvenile homicide and, 292 sexual homicide and, 343 Ashcroft, John, 231 Assessment. See also Diagnosis diagnostic labeling versus indepth psychiatric, 375–377 of homicidal behavior in schizophrenic individuals, 103–116 Atkins v. Virginia (2002), 411 Attachment theory borderline personality disorder and, 128, 129–130

HOMICIDE: A PSYCHIATRIC PERSPECTIVE psychotic depression and, 257 sexual homicide and, 337 Attention-deficit/hyperactivity disorder (ADHD), and juvenile homicide, 293, 295, 296, 312 Austria, and rates of homicide, 27 Authority figures, and victimology, 397 Automatisms, and relationship between seizures and violent behavior, 67, 69–70 Autonomic nervous system, and seizures, 67. See also Central nervous system Autonomy, and sexual homicide, 342. See also Control Avoidant personality disorder, 294 Barrow, Clyde, 44 Battered woman syndrome. See also Physical abuse interpersonal perspective on, 207–209 posttraumatic stress disorder and, 79 psychiatric considerations in, 209–210 society’s image of women and, 213–214 validity of defense in cases of, 214–215 Beck Depression Inventory, 305 Behavior, definition of homicidallevel, 245–246. See also Aggression; Impulsivity; Irritability; Paranoid behavior; Self-destructive behavior; Stalking behavior; Violence and violent behavior Behaviorism, and aggression, 56 Bender, Loretta, 302

Index Bereavement, and secondary victims of homicide, 414–415 Berkowitz, David, 36, 343–344 Bianchi, Kenneth, 36, 340 Bias, and hate crimes, 32–33 Biological disease model, of alcohol use, 24 Biological factors, in homicide. See also Genetics; Neurophysiology; Neurotransmitters; Physiological hypotheses borderline personality disorder and, 122, 124 dilemmas in perspectives on, 55–56 early theories on source of aggression and, 56–59 periodicity and, 27 problems with research studies on, 85–86 Biosocial approach, to homicidal violence, 86 Bipolar disorder cognitive impairments and violence in, 112 prevalence of violence and, 106 psychotic behavior and, 243 rationality and, 389 reassurance and, 253 Blame, of others legal concept of individual responsibility and, 382 narcissistic personality disorder and, 180 Blameworthiness, as legal concept, 201 Borderline pathology of childhood, 126 Borderline personality disorder depression and, 241 epidemiology of violent behavior in, 130–132

421 etiology of and propensity toward violence, 123–130 juvenile homicide and, 296–299 legal considerations and personal responsibility in, 146–151 lying or deception in, 370–371 masochism and, 224 narcissistic personality disorder and, 197 predisposition to violence in, 132–146, 407 strain theory and, 32 validity of diagnosis of, 121–122 Bowlby, John, 129 Brady, Ian, 346 Branch Davidians (Texas), 45–46, 150, 258 Brief Psychiatric Rating Scale, 111 British Medicines and Healthcare Products Regulatory Agency, 272 Brittain, Robert P., 341 Brudos, Jerome, 350 Bullying, in schools, 320 Bundy, Ted, 36 Buono, Angelo, 36 Bureau of Alcohol, Tobacco and Firearms, 46 California Menendez case, 204 and studies of homicide, 247, 312 Tarasoff case, 94–95, 384 Canada, and studies of homicide, 105, 249, 266 Capital punishment effect of increased prevalence of, 411 as instrumentalized type of violence, 398

422 Capital punishment (continued) juvenile homicide and, 322 social control by use of violence and, 401 Capgras’ syndrome, 111, 258 Case examples, of homicide altered states of consciousness and, 73–75 borderline personality disorder and, 137–143 delusions and, 108–110 dependent personality disorder and, 159–160, 162, 166, 172 health care professionals and medical homicide, 233–235 juvenile homicide and, 297, 298, 301, 303–304, 308, 309 lying and, 371 masochism and, 226, 228–229 narcissistic personality disorder and, 189–191, 195–196 postpartum psychosis and, 267– 268, 269–270 psychotic depression and, 252– 253, 254, 262 schizophrenia and, 113–115 seizure disorders and, 70–71 sexual homicide and, 346–347, 350, 351–352, 353–354 voluntariness of conduct and, 386 Catathymic states psychotic depression and, 256– 257 sexual homicide and, 349, 352– 354, 355 Catatonic type, of schizophrenia, 103 Causality, and SSRI controversy, 275 Causal networks, and voluntariness, 382–383 Centers for Disease Control, 286

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Central nervous system conduct disorder in children and, 296 neurological examination of juveniles and, 312 seizures and, 67 Cerebral blood flow (CBF), and violent behavior, 84–85 Cerebral hemisphere dysfunctions, and physiological hypotheses of violence, 81–85 Child abuse. See also Physical abuse; Sexual abuse infanticide and, 265 juvenile homicide and, 289, 315– 318 sexual homicide and, 348 Child Behavior Checklist, 65 Children. See also Adolescents; Age; Child abuse; Development; Family; Filicides; Infants and infanticide; Juveniles; Parenting borderline personality disorder and, 126–127 dependent personality disorder and, 157–158 lying or deception by, 370 personality disorder diagnoses in, 292–293 secondary effects of homicide on, 414, 415 sexual homicide and, 346–347 Chronic catathymic homicides, 352– 353 Chronic fatigue syndrome, and dependent personality disorder, 168 Civil commitment legal requirements for and violent behavior, 248 public image of mentally ill and, 92 sexual offenders and, 357

Index Clearance rate, for homicides, 16 Clinical appraisals, and predictions of violence in mentally ill, 96 Cocaine, and violent crimes, 25, 26, 48, 49 Codependence, and dependent personality disorder, 158 Cognitive functioning, and schizophrenia, 113 Cognitive mapping and processing, and sexual homicide, 338 Colombia, and rates of homicide, 27 Columbine High School (Littleton, CO) school shooting (1999), 319–320, 321–322 Combe, George, 82 Command hallucinations, 108, 254, 383 Common law felony murder and, 33 intent and, 368 sleepwalking and, 72 Community, and public image of mentally ill, 92 Comorbidity, of mental disorders borderline personality disorder and, 122, 133, 139, 147–148 dependent personality disorder and, 167–168 depression and, 240–241 schizophrenia and, 98, 102, 104, 106 self-defeating personality disorder and, 216 “Compatibility” question, and free will, 381 Compensation awards, for crime victims, 414 Competency, and schizophrenia, 116 Complex partial seizures, and violence, 62–63, 64, 65, 66–67

423 Compulsive sexual homicide, 353, 355 Conduct disorder, and juvenile homicide, 294, 295, 300, 316, 347 Confessions, of serial killers, 39 Conflicts dependent personality disorder and, 158–163, 174 measurements of success in treating homicidal behavior and, 367 Consciousness, altered states of dissociative disorders and, 314 seizure disorders and, 66–67 voluntariness of conduct and, 72– 75 Context. See also Environmental factors epidemiology of homicide and, 17 of homicide in battered woman syndrome cases, 212 human interaction and relationships, 396–397 Control. See also Episodic dyscontrol masochism and, 224 narcissistic personality disorder and, 185, 197 sexual homicide and, 338, 342 Coping, with secondary effects of homicide, 415 Cormier, Bruno, 172–173 Count of Monte Crisco, The (Dumas), 193 Crichton, Michael, 63 Crime. See also Aggravated assault; Arrest rates; Arson; Homicide; Law enforcement; Sexual offenses; Victim epidemiology of homicide during commission of, 14, 33–35 epidemiology of homicide perpetrator’s prior, 18–22

424 Crime (continued) nature and background of juvenile, 288–289 relationship of other offenses to homicide, 412–413 “Crime of passion,” and narcissistic personality disorder, 188 Crime and Punishment (Dostoyevsky), 353 Criminology. See Victimology Crisis intervention, and perception of self as victim, 202 Cults, and mass homicidal violence, 257–258. See also Branch Davidians; Jones, James; Religion Culture. See also Social issues cross-cultural studies of schizophrenia and violence, 112 epidemiology of homicide and, 13, 27–28 honor and violence in, 194 terrorist mass killings and, 45 “Culture of control,” 402 “Cycle of violence,” 315 Cyproterone acetate (CPA), 78 Dahmer, Jeffrey, 36, 340, 342, 345–346 Dangerousness. See also Duty to warn; Prediction; Risk factors and risk assessment civil commitment and, 248 legal concept of imminent danger, 213 prediction of violence in mentally ill and, 95 self-destructive behaviors in borderline personality disorder and, 135 sexual offenders and, 356, 358 Daubert case, 274

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Decision making dependent personality disorder and, 167 psychotic depression and, 260– 262 Defenses. See also Denial; Diminished-responsibility defense; Insanity defense; Selfdefense; Splitting borderline personality disorder and, 144, 298–299 dependent personality disorder and, 163, 174, 175 Deinstitutionalization, and public image of mentally ill, 92 Delgado, Jorge, 93 Delusional major depression, 242 Delusions and delusional disorder depression and, 242–245 differential diagnosis of schizophrenia and, 103 juvenile homicide and, 302, 303– 304 masochism and, 224–225 narcissistic personality disorder and, 197 psychotic depression and, 254– 257, 260, 261–263 schizophrenia and violent behavior, 107–110, 115, 116, 302, 303–304 Demographics. See Age; Race; Socioeconomic status Denial. See also Defenses dependent personality disorder and, 167, 175 psychotic depression and, 260, 266 sexual homicide and, 348 Denmark, and studies of homicide, 248, 251 Department of Justice, 12

Index Dependent personality disorder (DPD) behavior patterns leading to violence in, 168–172 childhood and, 157 paradox of violence in, 157 perpetrators’ reactions following homicide, 173–174 predisposing factors in, 158–168 therapeutic concerns for, 174 Depersonalization. See also Dissociation dependent personality disorder and, 174 psychotic depression and, 258– 259 Depo-Provera (Medroxyprogesterone acetate), 78 Depression. See also Psychotic depression battered woman syndrome and, 209 borderline personality disorder and, 122, 133–134 dependent personality disorder and, 167, 171 diagnostic specificity and, 240– 245, 259–260 juvenile homicide and, 304–311 masochism and, 223 narcissistic personality disorder and, 180–181, 183 prevalence of psychotic features with delusions in, 112 prevalence of violent behavior in, 106 rationality and, 389 review of studies of homicide and, 245–250 strain theory and, 32

425 two-pronged approach to understanding of, 239–240 DeSalvo, Albert, 36 Determinism genetic theories on violence and, 61 legal versus clinical views of voluntariness and, 381–382, 384 Development. See also Attachment theory; Children child abuse and, 317–318 lying and deception, 370 narcissistic personality disorder and rage states, 184–186 personality disorders in children and, 294 sexual homicide and, 337–339 Diagnosis. See also Assessment; Differential diagnosis of borderline personality disorder, 121–122, 131–132 specificity of for depression, 240–245, 259–260 specificity of for schizophrenia, 103–106 use of psychiatric in legal system, 375–381, 413 of violent behavior in juveniles, 299–302 Diagnostic fallacy, in legal settings, 380–381 Diagnostic Interview Schedule (DIS), 249 Diagnostic manuals. See DSM-III; DSM-III-R; DSM-IV; DSM-IV-TR Differential diagnosis of delusional disorder and schizophrenia, 103 use of in legal settings, 379

426 Diminished-responsibility defense postpartum psychosis and, 270 SSRI type of defense and, 275 Disconnected quality of behavior, and borderline personality disorder, 144–146 Dismemberment cases, and sexual homicide, 344–346, 350, 353–354 Disorganized type, of schizophrenia, 103 Disruptive behavior, and personality disorders in juveniles, 291 Dissociation. See also Depersonalization altered states of consciousness and, 72 borderline personality disorder and, 145–146 juvenile homicide and, 313–315 narcissistic personality disorder and, 194–196 psychotic depression and, 244, 258, 259 Dissociative identity disorder (DID), 314 Dissociative trance disorder, 174 Domestic violence, and decline in homicide rates, 49. See also Battered woman syndrome Dominance, and sexual homicide, 339 Dopamine, and role of neurotransmitters in violence, 81, 124 Dothiepin, 271 Double depression, 134 Drug use. See Substance abuse DSM-III, and borderline personality disorder, 121 DSM-III-R borderline personality disorder and, 121

HOMICIDE: A PSYCHIATRIC PERSPECTIVE narcissistic personality disorder and, 184 schizoaffective disorder in, 103, 243 self-defeating personality disorder and, 217, 218, 219 trauma and, 385 DSM-IV narcissistic personality disorder and, 184 schizophrenia and, 103 trauma and, 385 DSM-IV-TR borderline personality disorder and, 132 conduct disorder and, 295, 300 dependent personality disorder and, 174 dissociative disorders and, 314 narcissistic personality disorder and, 177 personality disorders in children and, 293, 294 postpartum psychosis and, 266 premenstrual syndrome and, 75 schizoaffective disorder and, 243 schizophrenia and, 103 self-defeating personality disorder and, 219 traumas as defined in, 385 use of in legal settings, 377 Durkheim, Emile, 402–403 Duty to warn. See also Prediction antidepressants and violent behavior, 273 mandated-reporting legislation and, 206 prediction of violence in mentally ill and, 94 Dyscontrol syndrome, and seizures, 68 Dysthymia, 240

Index East, W. Norwood, 82 Economic theory, and decline in homicide rates, 49 “Ecstasy” (Methylenedioxymethamphetamine), 26, 49 Ego functioning. See also Superego dependent personality disorder and, 170–172 individual differences in violent behavior and, 407 Electroencephalograms (EEGs) neurological examinations of juveniles and, 312 seizure disorders and, 66, 67, 68, 70 El Salvador, and rates of homicide, 27 Empathy, narcissistic personality disorder and lack of, 178, 182, 184, 195 Endocrine diseases, and hormonal theories of violence, 77 Endogenous depression, 241–242 England gender and homicide rates in, 13 infanticide in, 270 “Inquiries After Homicide” required in, 396 juvenile violence and homicide in, 289, 292, 302, 311 murder-suicide combination in, 251 public opinion on capital punishment in, 398 sexual homicide in, 340 study of antidepressants and violence in, 271 Entitlement, sense of borderline personality qualities in adolescents and, 299 narcissistic personality components and, 179, 183

427 Environmental factors, and epidemiology of homicide, 26. See also Context Epidemiologic Catchment Area (ECA) study, 102, 247, 249 Epidemiology, of homicide age and, 3–4, 5–8, 19, 20–21 borderline personality disorder and, 130–132 culture and, 27–28 decline in rates of, 1, 2, 48–50 depression and, 245–250 felony murders and, 33–35 gender and, 1, 5–8, 12–13, 15, 15, 200 genetics of violence and, 61 historical perspective on, 1–3 juveniles and, 18–19, 285–290 masochism and, 200–201 multiple-victim murders and, 35–47 perpetrator’s prior criminal offenses and, 18–22 race and, 1–2, 4–11, 10–11, 13 social relationships and, 14–18 sociological variables and, 28–30 subculture of violence and, 30–33 substance abuse and, 22–26 temporal and ecological factors and, 26–27 Episodic dyscontrol juvenile violence and, 303 seizures and, 64 Episodic killers, health care professionals as, 233 Erotomania, 111 Escalating encounters, and violent situations, 409 Estonia, and rates of homicide, 28

428 Ethics. See also Morality; Values capital punishment and, 411 normative versus nonnormative arguments and, 370 Ethological theories, on aggression, 57–59 Etiology, of borderline personality disorder, 123–130 Euthanasia, 230. See also Mercy killings Evidence, homicides to destroy, 332, 333, 334 Expert testimony admissibility of in battered woman syndrome cases, 213 sexual offenders and, 356 Explicit versus vague approaches, in legal and clinical views of homicide, 371–372 Exploitation, and narcissistic personality disorder, 181 Expressive motivation, 17, 18 False memory syndrome, 307 Familicide mass killings and, 44 psychotic depression and, 253 Family. See also Children; Familicide; Family systems approach; Filicides; Marriage; Parenting; Parricide epidemiology of homicide and, 15 mass killings and, 44 secondary effects of homicide and, 414 Family studies, of violent behavior, 60 Family systems approach, to intergender violence, 211 Fantasy battered woman syndrome and, 208

HOMICIDE: A PSYCHIATRIC PERSPECTIVE dependent personality disorder and, 169 psychotic depression and, 255– 256 serial killings and, 41–42 sexual homicide and, 334, 338, 339, 341–342, 349, 350–351, 353, 354 Fatal Attraction (movie), 140 Federal Bureau of Investigation (FBI), 1, 37, 46, 100, 320–321, 335–340, 394 Federal Hate Crime Statistics Act (1990), 32 Feedback filter, and sexual homicide, 339 Felony murders, 33–35 Feminist critiques of disease status of premenstrual syndrome, 76 of film presentation of stalking behavior, 140 of problem of intergender violence, 211–212 Fenfluramine, and borderline personality disorder, 125 Feticide, 264 Filicides, and psychotic depression, 264–270 Finland, and juvenile homicide, 311 Firearms. See also Gun control epidemiology of homicide and, 12, 13–14, 48 juvenile homicide and, 284, 287, 290, 318–319 Fixed-action patterns, and aggression, 58–59 Fluoxetine, 271–272 Focal seizures. See Partial seizures Follow-up study, of suicide risk in borderline personality disorder, 134

Index Food and Drug Administration (FDA), 273 Formal functions, of legal versus psychiatric system, 372–375 France, and studies of homicide, 24 Freeman group (Montana), 45 Free will. See also Voluntariness determinist view of, 381 sexual behavior and, 342, 356 Freud, Sigmund, 57, 221 Frontal lobe dysfunctions, and violent behavior, 83, 84 Frustration, and narcissistic personality disorder, 184 GABA, and borderline personality disorder, 124 Gacy, John Wayne, 36, 38 Gage, Phineas, 82–83 Gall, Francis Joseph, 82 Gangs, and juvenile homicide, 19, 287, 404 Garland, David, 402 Gender alcohol and violent behavior, 25 battered woman syndrome and, 211 borderline personality disorder in children and, 126, 130 epidemiology of homicide and, 1, 5–8, 12–13, 15, 15, 200 juvenile homicide and, 287 secondary effects of homicide and, 415 violent situations and, 410 Generalized seizures, 63, 67 Genetics borderline personality disorder and, 124 ethological theories on aggression and, 57–58

429 violent behavior and, 59–62, 316, 317 Glaxo SmithKline, 273 Gonzalez, Juan, 93 Götterdämmerung finale, and juvenile homicide, 309–310 Government. See Legal issues and legal system; Politics; State Grandiosity borderline personality disorder in adolescents and, 299 delusions and hallucinations in bipolar disorder, 243 narcissistic personality disorder and, 177–179, 182, 185, 186, 195 Grief. See Bereavement; Loss Group for the Advancement of Psychiatry, 357 Group settings, and acts of suicide or homicide, 257–258 Group violence, and research on homicide, 403–404. See also Gangs Gun control, 48, 290. See also Firearms Hallucinations depression and, 242–245, 255 juvenile homicide and, 302, 303– 304 schizophrenia and violent behavior, 107–110, 116, 302, 303–304 Harris, Eric, 319, 321–322 Hate crimes, and strain theory, 32–33 Hatred borderline personality disorder and, 123, 133, 136, 137, 139– 141, 144 psychotic depression and, 261

430 Head injuries, and violent behavior, 83, 84, 406 Health care professionals, masochism and homicides by, 229–235 Helplessness depression in juveniles and, 304 learned helplessness, and battered women, 209 masochism and, 204–206 Hennard, George, 43 Hinckley, John, 225 Hindley, Myra, 346 History, of homicide rates from 1930s to present, 1–3 Holmes, Oliver Wendell, Jr., 194 Homeless, prevalence of mental illness among, 93 Homicidal-level behavior, use of term, 245–246 Homicide biological factors in, 55–86 borderline personality disorder and, 121–151 dependent personality disorder and, 157–175 depression and, 239–276 epidemiology of, 1–50 juveniles and, 283–322 legal versus clinical views on, 363–390 masochism and, 199–235 murder-suicide combination, 251 narcissistic personality disorder and, 177–197 research needs for twenty-first century, 393–417 schizophrenia and, 91–116 sexual forms of, 331–339 Honor, and violent cultures, 194

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Hopelessness juvenile homicide and, 304, 309 psychotic depression and, 255, 304 Hormonal theories, of violence, 75–79 Hospitals and hospitalization, psychiatric. See also Civil commitment prediction of dangerousness and, 95 research strategies and, 395–396 studies of mental disorders and violent behavior in, 246–247, 248 studies of schizophrenia and violence in, 98–99 Household activity ratio, 29 Huberty, James, 43 Hypnosis, and altered states of consciousness, 72–73 Hypoglycemic states, and physiological hypotheses for violence, 80 Hypothalamus, and rage reactions, 83 Hysteric dysphoric, 123 Iatrogenic homicides, and selective serotonin reuptake inhibitors, 270–276 Ibn-Tamas case (1979), 213 Identity, and borderline personality disorder, 128, 296–297. See also Self Immediate response, dependent personality disorder and need for, 166 Imminent danger, legal concept of, 212–213

Index Impulsivity. See also Irresistibleimpulse question borderline personality disorder and, 133, 137–139, 149 child abuse and, 317 serotonin and, 25, 125, 272 Infants and infanticide. See also Children; Feticide; Filicides epidemiology of homicide and, 4 postpartum psychosis and, 76, 261, 264–270 Insanity defense. See also Diminished-responsibility defense borderline personality disorder and, 147 postpartum psychosis and, 266 schizophrenia and, 112–116 seizure disorders and, 69–70 SSRI type of defense and, 275 Insight, narcissistic personality disorder and lack of, 179–180 Instinctual theories, on aggression, 56 Institute of Living (Connecticut), 132 Instrumental model, of homicide, 397–399 Instrumental motivation, 17, 18 Intelligence quotient, and juvenile homicide, 313 Intent, legal definition of, 368 Interactions. See also Relationships context of homicidal acts and, 396–397 escalating encounters and, 409 types of and epidemiology of homicide, 16–18 International Coalition for Drug Awareness, 276 Ireland, and murder-suicides, 251 Irresistible-impulse question, and postpartum psychosis, 270

431 Irritability, and child abuse or neglect, 317 Israel, and studies of homicide, 13 Jack the Ripper, 36, 349 Jamaica, and rates of homicide, 28 Japan, and rates of homicide, 27, 28 Jones, James, 149, 258 Jonesboro, AK school shooting (1998), 319 Justice, and legal system, 194, 364– 365 Juveniles and juvenile homicide. See also Adolescents; Age; Children adult judicial system and, 322 childhood abuse and, 315–318 depression and, 304–311 dissociative phenomena and, 313–315 epidemiology of homicide and, 18–19, 285–290 neuropsychiatric factors and, 312–313 personality disorders and, 291– 302 schizophrenic disorders and, 302– 304 school shootings and, 318–322 special issues in homicide cases involving, 283–284 substance abuse disorders and, 311–312 Kaczynski, Ted, 37 Kansas v. Crane, 358 Kant, I., 382 Kernberg, O., 122, 127, 185, 223, 294, 297 Kevorkian, Jack, 230 King v. Cogdon (1950), 73–75 Klebold, Dylan, 319, 321–322 Kohut, H., 184, 193–94

432 Koresh, David, 45–46, 150 Kraepelin, E., 256 Krafft-Ebing, R. V., 333 Labeling, diagnostic versus in-depth psychiatric assessment, 375–377, 380 Language, biological predisposition for development of, 59 Latvia, and rates of homicide, 28 Law enforcement, and orientation of research on homicide, 336, 394. See also Arrest rates; Crime; Prisons and jails Learned-helplessness model, of battered woman syndrome, 209 “Learned intermediary doctrine,” 274 Learning disabilities, and juvenile homicide, 296, 313 Learning theory, and narcissistic personality disorder, 184 Legal issues and legal system. See also Capital punishment; Civil commitment; Common law; Crime; Expert testimony; Law enforcement; Prisons and jails; Supreme Court amnesia and, 314 antidepressants and iatrogenic homicides, 270–276 battered woman syndrome and, 212–213 borderline personality disorder and personal responsibility, 146–151 civil commitment and, 248 civil liability in cases of psychotic depression, 263–276 clinical views on homicide compared to, 363–390 concept of felony homicide in, 35

HOMICIDE: A PSYCHIATRIC PERSPECTIVE concepts of justice and, 194, 364, 365 dependent personality disorder and, 165, 173, 174 duty to warn and, 94–95 genetics of violence and, 61–62 juvenile homicide and adult judicial system, 322 research on predisposition of homicidal individuals within, 410–413 schizophrenia and homicidal violence, 112–116 seizures and violent behavior in, 69–72 sexual homicide and, 355–359 social control through use of violence, 401 victimology and, 201 Legitimate use, of violence, 398–399, 400–403 Lifestyle, and risk for victimization, 28–30 Life-support systems, termination of, 231 Lithuania, and rates of homicide, 28 Lombroso, Cesare, 82 Loneliness borderline personality disorder and, 141–143 dependent personality disorder and, 169 Longitudinal studies, of violence in borderline personality disorder, 132 Loss, and secondary effects of homicide, 415 Lucas, Henry Lee, 36 Lust homicides, 333–334 Lying, legal versus clinical views of, 370–371 Lysergic acid diethylamide (LSD), 25

Index MacArthur Study of Mental Disorder and Violence, 106, 107, 108 Maier, Hans, 256, 352 Malignant masochism, 227–229 Malvo, Lee, 44 Mandated-reporting legislation, 206 Manic episodes. See also Bipolar disorder diagnosis of depression and, 240 psychotic behavior and, 243 rationality and, 389 Manipulation borderline personality disorder and, 123, 136, 138, 141, 144 narcissistic personality disorder and, 181 Manson, Charles, 149–150 Marriage. See also Battered woman syndrome; Family dependent personality disorder and homicide in context of, 172–173 secondary effects of homicide and, 414 Masochism. See also Sadomasochism abused person and homicide, 207–215 health care professionals and, 229–235 psychodynamic perspective on, 220–229 self-defeating personality disorder and, 215–220 victimology and, 199–207 Masochistic personality disorder, 216–217 Mass killings definition of, 35 familicide and, 44 medical mass murder and, 44 serial killings and, 42

433 spree killers and, 44 terrorism and, 45–47 McVeigh, Timothy, 43 MDMA (Methylenedioxymethamphetamine), 26, 49 Medea (Euripides), 268–269 Media borderline personality disorder and, 132 exposure to violence and, 405 juvenile homicides and, 284, 289, 320 schizophrenia and, 91 school shootings and, 320 secondary effects of homicide and, 415 serial killers and, 37 sexual homicide and, 355 violent situations and, 409–410 Medical care documentation of seizure disorders and, 70 homicides by health care professionals and, 229–235 improvement in and decline in homicide rates, 49 Medical liability, and cases of psychotic depression, 264 Medical mass murder, 44 Medroxyprogesterone acetate (Depo-Provera), 78 Melancholic depression, 244–245 Memory, and borderline personality disorder, 145 Menendez case (California), 204 Mens rea (criminal mind or intent), 69, 368, 369 Menstrual cycle, and hormonal theories of violence, 76–77 Mental abnormality, legal definition of, 357

434 Mental retardation, and capital punishment, 411 Mercy killings dependent personality disorder and, 171 euthanasia, 230 medical mass murders and, 44 Merging, and narcissistic personality disorder, 179 Merton, R., 32 Methamphetamine, 26, 49 Methods of killing, and epidemiology of homicide, 12–13, 13–14. See also Firearms; Weapons Methylenedioxymethamphetamine (MDMA), 26, 49 Michael Kohlhaas (Heinrich von Kleist), 193–194 Michigan, and physician-assisted suicide, 230 Military, and group violence, 404, 408 Minnesota, and psychopathic personality statute, 356–357 Minnesota Multiphasic Personality Inventory, 305 Mob violence, 403–404 Moby Dick (Melville), 193 Model Penal Code, 71, 368 Molecular genetics, 59 Mongolia, and rates of homicide, 27 Monoamine oxidase A (MAO-A) gene, 316, 382 Montalvo Puente, Dorothea, 36 Mood disorders, overlap of borderline personality disorder with, 122 Moon, and correlation of lunar phase cycle and homicides, 27

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Morality. See also Ethics; Values prohibitions against killing and, 370 treatment of borderline personality disorder and, 151 Motivations, for homicide instrumental versus expressive, 17, 18 sexual homicide and, 336–339 Motus v. Pfizer case, 273 Muhammad, John, 44 Multiple complex developmental disorder, 297 Multiple-victim murders, epidemiology of, 35–47 Murder-suicide combination, and psychotic depression, 251 Mutuality of Autonomy Scale, 134 Myths, about school shootings, 3 21 Narcissism borderline personality disorder in juveniles and, 299 masochism and, 222–223, 224 sexual homicide and, 343 Narcissistic personality disorder characteristics of, 177–181 psychodynamics of, 196–197 rage and homicidal violence in, 181–196 National Academy of Sciences, 1 National Center for the Analysis of Violent Crime, 37, 342 National Center for Health Statistics, 286 National Health Service (England), 396 Nazi Germany, and homicides by physicians, 230, 233 Neonatal mortality, 264, 265

Index Neurology. See also Neuropsychological testing evaluations of juveniles and, 312– 313 schizophrenia and violent behavior, 104–105, 113 Neurophysiology. See also Biological factors; Neurotransmitters; Physiological hypotheses borderline personality disorder and , 125 determinism and, 382 differences among individuals and, 406–407 rage reactions in narcissistic personality disorder and, 184 Neuropsychiatric contributions, to juvenile homicide, 312–313 Neuropsychological testing, and seizure disorders, 66 Neurotic depression, 241–242, 251 Neurotransmitters. See also Serotonin borderline personality disorder and, 124–125 physiological hypotheses on violence and, 80–81, 124, 125, 272, 406 New York, and studies of homicide, 4, 30, 249–250, 289, 312 New York Psychiatric Institute, 132 Nihilism, and psychotic depression, 259 Nilsen, Dennis, 344–345 Noradrenaline, and role of neurotransmitters in violence, 81 Normative versus nonnormative perspectives, on homicidal behavior, 369–371 Norway, and rates of homicide, 28

435 Object representations, and borderline personality disorder, 127, 144 Obsessional symptoms concepts of rationality and, 389 dependent personality disorder and, 167 Obsessive-compulsive disorder, and borderline personality disorder, 128–129 Ohio, and juvenile crime, 288 Oklahoma City bombing (1995), 43 Oppositional defiant disorder (ODD), 294 Oregon, and physician-assisted suicide, 230 Organized/disorganized dichotomy, in sexual homicide, 339–340 Othello syndrome, and dependent personality disorder, 161 Paranoid behavior concepts of rationality and, 389 narcissistic personality disorder and, 181 Paranoid delusions, 243 Paranoid schizophrenia diagnosis of, 103 parricides and, 105 state of mind during crime and, 110 Parenting, and juvenile homicide, 317 Parker, Bonnie, 44 Paroxetine, 271, 273, 274–275, 276 Parricide, and juvenile homicide, 303–304, 308, 316 Partial seizures, 63, 64 Pearl, MS, school shooting (1997), 319

436 Peer groups, and juveniles with borderline personality disorder, 298 Pennsylvania, and studies of homicide, 4, 289 Perinatal death, and concept of feticide, 264 Periodicity, and biological factors in homicide, 27 Persecutory delusions, 107 Personality. See also Narcissism of health care professionals committing homicides, 232 seizure disorders and interictal, 64–65, 67, 68 Personality Disorder Examination, 219 Personality disorders. See also Antisocial personality disorder; Borderline personality disorder; Dependent personality disorder; Narcissistic personality disorder diagnosis of in children, 292–293 juvenile homicide and, 291–302 masochism and self-defeating personalities, 215–220 overlap of borderline personality disorder with other, 122 overlap of dependent personality disorder with other, 167–168 psychiatric diagnoses in legal settings and, 379–380 schizophrenia and comorbidity with, 104 voluntariness and, 383 Pfizer Inc., 273, 274 Phrenology, and theories of violence, 82

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Physical abuse. See also Battered woman syndrome; Child abuse; Sexual abuse borderline personality disorder and, 148 hospital studies and, 395 juvenile homicide and, 289, 306– 307 self-defeating personality disorder and, 217 sexual homicide and, 337, 355 Physician-assisted suicide, 230–231 Physiological hypotheses, on violence. See also Biological factors; Neurophysiology altered states of consciousness and, 72–75 borderline personality disorder and, 124–125 cerebral hemisphere dysfunctions and, 81–85 hormonal theories and, 75–79 neurotransmitters and, 80–81 posttraumatic stress disorder and, 79–80 seizure disorders and, 62–72 Pinel, Philippe, 82 Pittman case (South Carolina), 275– 276 Politics. See also State capital punishment and, 411 group violence and, 403 homicides by health care professionals and, 233 mass killings and, 43, 47, 408 serial killings of political prisoners and, 38 Pornography, and sexual violence, 410 Positivistic criminology, 82 Post, Jerrold, 45

Index Postpartum psychosis, and infanticide, 76, 261, 264–270 Posttraumatic stress disorder (PTSD) battered woman syndrome and, 209–210, 214 psychiatric diagnoses in legal settings and, 379–380 self-defeating personality disorder and, 216, 219–220 voluntariness of conduct and, 79– 80, 385, 386 Power dependent personality disorder and, 159 narcissistic personality disorder and, 182 serial killers and, 39 sexual homicide and, 342 Prediction, of violence. See also Dangerousness; Duty to warn; Rick factors and risk assessment diagnostic classifications in legal settings and, 413 schizophrenia and, 94–97 tables for, 412–413 Predisposition, of individual to violence, 132–146, 404–405, 407 Prefrontal cortex, and violent behavior, 85, 242 Pregnancy. See Postpartum psychosis Premenstrual syndrome (PMS), and hormonal theories of violence, 75–77 Prevalence of comorbidity in borderline personality disorder, 133 of depression, 245 of mental illness in homeless, 93 of psychotic depression, 244

437 of PTSD after commission of homicide, 79 of weapons carrying by adolescents, 290 Primary homicides, 14–15 Prisons and jails. See also Crime; Law enforcement; Legal issues and legal system increase in incarcerated population and, 48 narcissistic personality disorder and, 182 prevalence of mental disorders in, 247, 249 rate of imprisonment and, 402 sentencing trends and, 412 Probability theory, 96 Projective identification, and psychotic depression, 257 “Prozac defense,” 273 Pseudologia fantastica, 145, 370–371 Pseudopsychopathic schizophrenia, 302 Psychiatric Epidemiology Interview, 249 Psychiatric examinations, mandatory for individuals accused of homicide, 247–248 Psychoanalytic theory on aggression, 57 on borderline personality disorder, 127–129 Psychodynamic theory on masochism, 220–229 on narcissistic personality disorder, 184, 196–197 on psychotic depression, 251–259 on sexual homicide, 348–352, 354– 355 Psychological autopsy, after suicide, 409

438 Psychological testing, in schizophrenia, 113 Psychomotor agitation, and psychotic depression, 244 Psychomotor retardation, and psychotic depression, 254 Psychomotor seizures, 65 Psychosis and psychotic symptoms. See also Psychotic depression; Schizophrenia borderline personality disorder and, 149 juvenile homicide and, 297 seizure disorders and, 68 Psychosurgery, and seizure disorders, 62–63 Psychotic depression. See also Depression delusions and, 254–257, 260, 262– 263 diagnosis of, 259–260 factors triggering decision to act, 260–262 legal issues and, 263–276 prevalence of, 244 psychodynamic hypotheses on, 251–259 recognition of as distinct syndrome, 250 suicide and, 251 PTSD. See Posttraumatic stress disorder Public health, and epidemiology of homicide, 9 Public opinion capital punishment and, 398 juvenile homicides and, 284 perceptions of mental illness, 245 schizophrenia and, 92–93

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Punishment, and violent behavior of individuals, 405–406. See also Capital punishment; Prisons and jails; Rehabilitation Quill, Timothy, 230 Race epidemiology of homicide and, 1– 2, 4–11, 10–11, 13 juvenile homicide and, 287, 288 postpartum psychosis and, 269 suicides of adolescents and, 286 Rage. See also Anger borderline personality disorder and, 139–141, 151 narcissistic personality disorder and, 181–196, 197 Ramirez, Richard, 36 Rape homicides, 333 Rational choice model, 372 Rationality, legal versus clinical views of, 387–390 Rawls, John, 194 Reactive depression, 241 Reality borderline personality disorder and, 143, 297 psychotic depression and, 258– 259, 260 secondary effects of homicide and, 414 Real-offense sentencing, 412 Recidivism measurements of success and, 366 of sexual offenders, 356, 357, 358– 359 Reconviction rates legal versus clinical views, 366– 367

Index relationship of other offenses to homicide and, 412 Red Lake Indian Reservation (Minnesota) school shooting (2005), 319 Regressions, and borderline personality disorder, 127 Rehabilitation criminal justice system and, 364 of sexual offenders, 356, 357 Rejection, and masochism, 226 Relationships. See also Family; Interactions; Marriage borderline personality disorder and, 128, 129, 130, 131, 136, 137, 138, 141–143 dependent personality disorder and, 157–175 epidemiology of homicide and, 14–18, 15 formal functions and definitions of, 373, 374–375 schizophrenia and homicidal violence, 104 serial killers and, 39 therapeutic intervention and disruption of, 205 victimology and, 396–397 Reliability, of borderline personality disorder diagnosis, 122 Religion. See also Cults; Politics; Rituals group violence and, 403 terrorist mass killings and, 45, 46– 47 Remorse, narcissistic personality disorder and lack of, 195–196 Repression, and dependent personality disorder, 175

439 Research, future of on homicide and violence on group violence leading to homicide, 403–404 on nature and extent of homicide in contemporary society, 394–396 on predisposition of homicidal individuals within legal system, 410–413 on problem of risk assessment, 416 on role of instrumental violence in homicide, 397–399 on secondary effects of homicide, 414–415 on social control and violence, 400–403 on social meanings of violence, 399–400 on victimology, 396–397 on violent individuals, 404–409 on violent situations, 409–410 Responsibility, borderline personality disorder and legal, 146–151. See also Diminishedresponsibility defense; Insanity defense; Voluntariness Retributive justice, 364–365 Retrospective approach, to study of schizophrenia and violence, 101–102 Revenge narcissistic personality disorder and need for, 191–194, 196 school shootings and, 320 Ridgway, Gary, 340 Rifkin, Joel, 340

440 Risk factors and risk assessment. See also Dangerousness; Prediction changes in lifestyle and, 28–30 dependent personality disorder and, 158–168 lifetime for homicide, 1 need for research on, 416 predictions about dangerousness and, 97 sexual offenders and, 358–359 Rituals serial killers and, 40 sexual homicide and, 355 Robbery homicides, 34 Roe v. Wade (1973), 49 Role models, and sexual homicide, 338 Roper v. Simmons, 322 Russia, and rates of homicide, 28 Sadism, and sexual homicide, 341– 342 Sadomasochism. See also Masochism juvenile homicides and, 307 sexual homicide and, 355 Schizoaffective disorder diagnostic classification of, 103, 243 parricides and, 105 Schizoid personality disorder, 347 Schizophrenia assessment of homicidal behavior in, 103–116 early studies of violence and, 97– 102 juvenile homicide and, 302–304 prediction of violence and, 94–97 prevalence of in prison populations, 247 problem of determining relationship between homicide and, 91

HOMICIDE: A PSYCHIATRIC PERSPECTIVE public image of, 91, 92–93 seizure disorders and symptoms of, 68 Schizotypal personality disorder, 347 School shootings, and juvenile homicide, 318–322 Scotland, and studies of homicide, 24 Seasonal factors, in incidence of violence, 26 Secondary effects, of homicide, 414– 415 Seizure disorders, and violent behavior, 62–72, 406 Selective serotonin reuptake inhibitors (SSRIs), and iatrogenic homicides, 270–276 Self. See also Identity; Self-esteem borderline personality disorder and, 130 masochism and perception of as victim, 202–204 narcissistic personality disorder and, 187 sexual homicide and, 349 Self-defeating personality disorder, 215–220, 222 Self-defense battered woman syndrome and, 208, 210, 212 prohibitions against killing and, 370 Self-destructive behaviors borderline personality disorder and, 135 malignant masochism and, 227–228 Self-esteem dependent personality disorder and, 163, 170 narcissistic personality disorder and, 178, 179, 181, 182, 183, 185, 187, 189, 192 psychotic depression and, 255

Index Self psychology, and narcissistic personality disorder, 186–187 Self-report studies, of homicide, 395 Separation-individuation process, and borderline personality disorder in juveniles, 298–299 September 11, 2001, terrorist attack, 43, 45, 46 Serial killers epidemiology of homicide and, 35–42 medical settings and, 233 sexual homicide and, 343 Serotonin. See also Neurotransmitters conduct disorder in children and, 296 impulsivity and, 25, 125, 272 relationship between alcohol and violence and, 25 Sertraline (Zoloft), 274, 276 Setting, and epidemiology of homicide, 14. See also Context; Group settings Sexual abuse. See also Child abuse; Physical abuse borderline personality disorder and, 148 juvenile homicide and, 289, 307 self-defeating personality disorder and, 217 sexual homicide and, 337, 355 Sexual homicide catathymic and compulsive types of, 349, 352–354, 355 children and, 346–347 classification of crime as, 331–332 descriptive models of, 334–344 dismemberment cases and, 344– 346, 350, 353–354 legislative and judicial responses to, 355–359

441 psychodynamics of, 348–352, 354– 355 taxonomy of, 332–334 Sexuality, and borderline personality disorder, 298 Sexually violent predators (SVPs), 357–358 Sexual offenses, and surgical or hormonal castration, 78 Shame, and narcissistic personality disorder, 189–192, 196, 197 Shawcross, Arthur, 340 Sherrill, Patrick Henry, 43 Sick role, and borderline personality disorder, 150–151 Side effects, of selective serotonin reuptake inhibitors, 271 Sign stimulus, and aggression, 58 Situated transaction, homicidal outcome as, 16–17 60 Minutes (television), 230 Sleepwalking, and altered states of consciousness, 72, 73–75 Smith, Susan, 269 Social-cognitive model, of alcohol use, 24 Social impact theory, and suicides or homicides in group settings, 258 Social issues. See also Culture; Relationships battered woman syndrome and image of women in, 213–214 research on meanings of violence and, 399–400 seizures and violent behavior as, 69–72 Socialization, and juvenile homicide, 288 Social psychology, and victim role, 203

442 Social stigma, and diagnosis of personality disorders in children, 293 Societal protection, legal versus clinical views on, 364–366 Socioeconomic status epidemiology of homicide and, 9, 29–30 juvenile homicide and, 313 postpartum psychosis and, 269 secondary effects of homicide and, 415 Sociology and sociological approaches battered woman syndrome and, 211–212 epidemiology of homicide and variables of, 28–30 juvenile homicide and, 301–302 masochism and victim role, 206– 207 need for joint psychiatric research on violence, 400 relationship between victim and killer and, 201 research on law and social control through use of violence, 401 Somatic delusions, and psychotic depression, 244, 259, 262 Somatization, and dependent personality disorder, 168 South (U.S.) capital punishment and, 411 honor and violence in culture of, 194 trends in homicide rates and, 3 South Africa, and rates of homicide, 28 Spain, and rates of homicide, 28 Specificity. See Diagnosis Speck, Richard F., 43, 60

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Splitting. See also Defenses borderline personality disorder and, 144, 299 sexual homicide and, 348, 349– 352 Spree killers, 35, 44 SSRIs (selective serotonin reuptake inhibitors), and iatrogenic homicides, 270–276 Stalking behavior borderline personality disorder and, 140–141 dependent personality disorder and, 169 masochism and, 225–227 schizophrenia and, 105 Starkweather, Charles, 44 State, and legitimate uses of violence, 400–403. See also Politics State of mind, and relationship between schizophrenia and violence, 110–112 Statistical-actuarial methods, for prediction of violence, 96–97 Stereotype, of battered women, 207– 208, 213 Strain theory, and subculture of violence, 31–33 Stress, and redefining of relationships, 374 Structural theory, and aggression, 57 Structured Clinical Interview for DSM-III-R, 219 Substance abuse borderline personality disorder and, 142–143, 147 conduct disorder in children and, 295 dependent personality disorder and, 161, 170

Index differences in violent behavior of individuals and, 408–409 epidemiology of homicide and, 22–26 hormonal theories of violence and, 77 juvenile homicide and, 311–312 schizophrenia and, 104, 106 Subtypes, of borderline personality disorder, 123 Success, measurements of in dealing with homicidal individuals, 366–367 Sudden murder, 255 Suicide and suicidal behavior adolescents and, 272–273, 274, 286, 309 antidepressants and, 271–272, 274 borderline personality disorder and, 130, 132, 134–135, 136 dependent personality disorder and, 161, 163–164, 171, 173 masochism and, 226–227 murder-suicide combination, 251 physician-assisted, 230–231 psychological autopsy after, 409 psychotic depression and, 251 serotonin hypothesis and, 125 Superego, and borderline personality disorder, 144, 299. See also Ego functioning Superpredators, 19 Supplementary Homicide Reports (SHRs), 1, 14, 17, 332 Supreme Court, 49, 61, 230, 322, 356– 357, 411 Swango, Michael, 233

443 Sweden, and studies of homicide, 28, 247–248 Switzerland, and rates of homicide, 28 Tantrums, and borderline personality disorder, 139–141, 151 Tarasoff case (California), 94–95, 384 Temporal lobe seizures, 63–66 Temporary insanity, and rage reactions in narcissistic personality disorder, 188 Terminal Man, The (Crichton), 63 Terrorism mass killings and, 45–47 political motivation and, 43 research on group violence and, 403 Testosterone, and hormonal theories of violence, 77–78, 317 Therapeutic interventions, and masochism, 205 Thinking and thought disorders, and psychotic depression, 252, 254– 256, 260 Third parties, as victims of homicidal violence, 160–161 Three-phase cycle, of violence, 208– 209 Time of day, and statistics on homicide, 26–27 Timing, of homicide in battered woman syndrome, 212 Tobin v. SmithKline Beecham (Wyoming), 274, 276 Torture, and serial killings, 38 Transference, and homicides by health care professionals, 232

444 Traumas borderline personality disorder and childhood, 129–130 posttraumatic stress disorder and, 79 sexual homicide and, 337, 348 Twinkie defense, 382 Twin studies, of violent behavior, 60, 124, 316 Uniform Crime Reporting (UCR) Program and Uniform Crime Reports, 1, 2, 3, 13, 14, 15, 26, 100, 331–332, 394 U.S. Secret Service, 320 U.S. Sentencing Commission, 32 Unworthiness, and narcissistic personality disorder, 177–179 Vacco v. Quill, 230 Validity. See Diagnosis Values. See also Ethics; Morality psychotic depression and delusions, 263 subculture of violence and, 31, 33 Victim. See also Relationships; Victimology adolescents as, 19, 20 definition of, 199–200 epidemiological data and, 3, 19 lifestyle and, 28–30 narcissistic personality disorder and role of, 181 perpetrator of homicide and role of, 200 research on homicide and surveys of, 395 Victimology future research and, 396–397 masochism and, 199–207 Victim-precipitation model, 16

HOMICIDE: A PSYCHIATRIC PERSPECTIVE Videotapes, and documentation of seizure disorders, 70 Vietnam War, and PTSD, 79 Violence and violent behavior. See also Research belief in alcohol as cause of, 23 borderline personality disorder and, 123–146 controversy on juvenile, 284 definitional problems and, 97–98 delusions and hallucinations, 107–110 dependent personality disorder and, 168–172 diagnosis of in juveniles, 299–302 early studies of schizophrenia and, 97–102 formal functions in dealing with, 374 genetics and, 59–62 physiological hypotheses for, 62– 85 popular image of mental illness and, 92 schizophrenia and prediction of, 94–97 SSRI medications and, 271–276 subculture of, 30–33 three-phase cycle of, 208–209 Voluntariness. See also Free will; Responsibility altered states of consciousness and, 72–75 autonomy, in sexual homicide, 342 legal versus clinical views of, 381– 387 seizure disorders and, 70, 71–72 Wagner, Ernst, 108–110 Washington, and weapons in high schools, 290

445

Index Washington v. Glucksberg, 230 Weapons, and juvenile crime, 289– 290. See also Firearms Wertham, Fredric, 256, 352 West Paducah, KY, school shooting (1997), 319 Whitman, Charles, 43 Williams, Andy, 314–315 Williams, Wayne, 36, 38

Workforce, and women as homicide victims, 200 Wuornos, Aileen, 36 XYY-violence hypothesis, 60, 61–62 Yates, Andrea, 267–268 Zoloft (sertraline), 274, 276 Zolpidem, 274

Homicide: A Psychiatric Perspective 2nd Edition

Metabolic Regulation: A Human Perspective, 2nd Edition

Human Nutrition: A Health Perspective, 2nd edition

Soviet Politics in Perspective: 2nd Edition

Psychiatric Rehabilitation, Second Edition

Organic Pollutants: An Ecotoxicological Perspective, 2nd Edition

Communicable Disease Epidemiology and Control: A Global Perspective, 2nd Edition

A Wee Christmas Homicide

A Wee Christmas Homicide

Homicide

Homicide

A Wee Christmas Homicide

Homicide: A Sociological Explanation

The American Psychiatric Press Textbook of Geriatric Neuropsychiatry, 2nd Edition

Homicide: A Sociological Explanation

A Presocratics Reader, 2nd Edition

Engineering a Compiler (2nd Edition)

Double homicide

Homicide Sanitarium

Homicide Johnny

Homicide Trinity

Double Homicide

Homicide Trinity

Small Homicide

Stress, Coping, and Development, 2nd Edition: An Integrative Perspective

Double Homicide

Homicide Trinity

Homicide Sanitarium

Homicide Hunch

The Case for Palestine: An International Law Perspective, 2nd edition

Double Homicide

Homicide: A Psychiatric Perspective 2nd Edition

Metabolic Regulation: A Human Perspective, 2nd Edition

Human Nutrition: A Health Perspective, 2nd edition

Soviet Politics in Perspective: 2nd Edition

Psychiatric Rehabilitation, Second Edition

Organic Pollutants: An Ecotoxicological Perspective, 2nd Edition

Communicable Disease Epidemiology and Control: A Global Perspective, 2nd Edition

A Wee Christmas Homicide

A Wee Christmas Homicide

Homicide

Homicide

A Wee Christmas Homicide

Homicide: A Sociological Explanation

The American Psychiatric Press Textbook of Geriatric Neuropsychiatry, 2nd Edition

Homicide: A Sociological Explanation

A Presocratics Reader, 2nd Edition

Engineering a Compiler (2nd Edition)

Double homicide

Homicide Sanitarium

Homicide Johnny

Homicide Trinity

Double Homicide

Homicide Trinity

Small Homicide

Stress, Coping, and Development, 2nd Edition: An Integrative Perspective

Double Homicide

Homicide Trinity

Homicide Sanitarium

Homicide Hunch

The Case for Palestine: An International Law Perspective, 2nd edition

Double Homicide

Recommend Documents